JP4370720B2 - Light-receiving element inspection apparatus and inspection method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an inspection device and an inspection method for a light receiving element that outputs an electrical signal corresponding to received light, for example.
[0002]
[Prior art]
A light receiving element such as a photodiode is used for CD (Compact Disc), CD-R (Compact Disc Recordable), DVD (Digital Versatile Disc) and the like.
For example, when reproducing a CD, laser light emitted from a semiconductor laser is reflected by a reflective film of the CD, and the reflected light is intensity-modulated by a depression of several μm in length called a pit on the reflection surface, The reflected light is converted into an electric signal corresponding to the reflected light by a light receiving element such as a photodiode, and the electric signal is converted into a digital signal to reproduce the signal.
[0003]
FIG. 19A shows a cross-sectional view of a semiconductor chip having a light receiving element used for the above-mentioned CD, for example.
As shown in FIG. 19A, the semiconductor chip 100 is formed with a light receiving element 101 made of, for example, a photodiode.
The semiconductor chip 100 is mounted on a lead frame 103. A pad (not shown) connected to the output terminal of each light receiving region of the light receiving element 101 formed on the semiconductor chip 100 and the lead frame 103 are made of a gold wire or the like. Connected by wires.
Further, a transparent resin 102 is formed so as to cover the periphery of the semiconductor chip 100, and a part of the lead frame 103 protrudes from the transparent resin 102 and can be connected to an external terminal.
The transparent resin 102 has a role of transmitting reflected light from the disk to the light receiving element.
[0004]
FIG. 19B is an enlarged plan view of the light receiving element 101 formed on the semiconductor chip 100 of FIG.
As shown in FIG. 19B, the light receiving element 101 has a rectangular light receiving region (A to F) divided into six. In this specification, the light receiving region is referred to as a pad.
For example, a predetermined electric signal (SA to SD) corresponding to the light amount is output from the light amount detected by the pads (A to D) of the light receiving element 101, and a predetermined electric signal (SA to SD) is output. By performing the calculation, a focus error signal is generated.
Further, for example, a predetermined electric signal (SE, SF) corresponding to the light amount is output from the light amount detected by the pads (E, F) of the light receiving element 101, and a predetermined calculation is performed on the electric signal (SE, SF). As a result, a tracking error signal is generated.
[0005]
In the inspection of the light receiving element 101 such as the photodiode described above, it is necessary to select a defect of several microns that is the effective diameter of the laser light within the surface of each pad (A to F).
As the inspection of the defect, for example, inspection of whether or not the transparent resin 102 on the semiconductor chip 100 is peeled off, whether or not there is a foreign substance in the transparent resin 102, and whether or not there is dirt is performed.
Further, for example, it is necessary to guarantee a deviation in sensitivity within ± 5% within the plane of each pad (A to F), between pads (A to D), and between pads (E to F).
[0006]
For the above guarantee, conventionally, in inspection of a light receiving element, a defect of several microns is picked up by an image pickup device by visual observation with a microscope or in a state where light is applied to the light receiving element by an illumination device, and image processing is performed. The image data is binarized based on brightness and darkness, and the presence or absence of defects in the light receiving element is inspected.
[0007]
[Problems to be solved by the invention]
However, in the above-mentioned visual inspection of humans, there is a problem that the level of inspection is not stable due to variation in experience time and human judgment, resulting in a decrease in yield and outflow of defective products. In the inspection by processing, there is a problem that the level of binarization depends on the illumination conditions.
[0008]
In addition, in either method, in-plane inspection is possible to some extent, but there is a problem that it cannot cope with inter-surface inspection.
Further, in the above conventional method, the presence or absence of peeling of the transparent resin 102 on the light receiving element 101, particularly when the transparent resin 102 on the light receiving element 101 is completely peeled off, is identified as a defective product. It was virtually impossible.
[0009]
SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems. Therefore, the present invention improves the accuracy and stability of inspection and can provide a light receiving element with high reliability and an inspection method for the light receiving element. The purpose is to provide.
[0010]
[Means for Solving the Problems]
  In order to achieve the above object, the light receiving element inspection apparatus of the present invention comprises:pluralA light receiving element that includes a light receiving area and outputs a light receiving signal corresponding to the light irradiation level to the light receiving areaTheInspectionTo doIrradiating the light receiving area of the light receiving element with a predetermined level of inspection light and scanning the inspection light on the light receiving areaPartData processing including a processing system that creates image data and performs predetermined processing based on a light reception signal from the light receiving element at each scanning point of the inspection lightPartAnd the data processingPartThe processing system binarizes the image data, extracts the shape of the light receiving region based on the binary image data, determines a reference position, and sets an inspection range based on the reference position A range setting unit; and an inspection unit that inspects defects of the light receiving element based on the image data within the inspection range set in the inspection range setting unit.The inspection range setting unit is configured to remove the dead zone area between the first light receiving area and the second light receiving area facing each other included in the extracted light receiving areas. Each image data of the first light receiving area and the second light receiving area is enlarged or reduced to create one apparent light receiving area, and the shape of each light receiving area after removing the dead zone is preset. The shape of each light receiving area of the set pattern is compared, and the reference position is determined by the light receiving area having the most correlation.
[0011]
  Preferably,The inspection range setting unit includes the first light receiving region so that the apparent area of the entire light receiving region is equal to the sum of the areas of the first light receiving region and the second light receiving region. The image data of the area and the second light receiving area is enlarged or reduced.
[0012]
  Preferably, the inspection range setting unit calculates four circumscribing sides circumscribing each other at a predetermined angle circumscribing the light receiving area having the most correlation, and circumscribing formed by the four circumscribing sides. The center of gravity of the rectangleAboveDetermine as the reference position.
[0014]
For example, the inspection unit determines the quality of the light receiving element based on the area of the shape of the light receiving region extracted based on the binary image data in the inspection range.
[0015]
For example, the inspection unit determines pass / fail of the light receiving element based on a density value corresponding to the strength of the light reception signal at each scanning point within the inspection range of the image data.
[0016]
For example, the inspection unit calculates an average value of the density values within the inspection range of the image data, and determines the quality of the light receiving element using the average value.
In this case, for example, the inspection unit compares the average value with a predetermined set value to determine the quality of the light receiving element.
Further, for example, the inspection unit determines pass / fail of the light receiving element based on a deviation from the average value of density values in the inspection range.
[0017]
  Preferably,eachThe data processing further includes a calculation unit that calculates a light reception signal corresponding to the light reception region.PartGenerates the image data of the light receiving element from the light reception signal calculated by the calculation unit.
[0018]
  According to the above-described light receiving element inspection apparatus of the present invention, light irradiationPartData processing based on the received light signal from the light receiving element at each scanning point of the inspection light fromPartThe image data is created byPartThe inspection range setting unit binarizes the image data, extracts the shape of the light receiving area based on the binary image data, determines the reference position, and sets the inspection range based on the reference position. .
  And data processingPartIn the inspection section, inspection of defects of the light receiving element is performed based on the image data within the inspection range set in the inspection range setting section.
  By setting the inspection range as described above and inspecting the defect of the light receiving element from the image data within the inspection range, it is possible to easily inspect the defect of the light receiving element from the image data, and Inspection can be performed with high accuracy.
[0019]
  Furthermore, in order to achieve the above object, the light receiving element inspection method of the present invention comprises:pluralA light receiving element that includes a light receiving area and outputs a light receiving signal corresponding to the light irradiation level to the light receiving areaTheInspectionTo doIrradiating the light receiving region of the light receiving element with a predetermined level of inspection light and scanning the inspection light on the light receiving region, and based on a light receiving signal from the light receiving element at each scanning point of the inspection light A step of creating image data, binarization processing of the image data, extraction of the shape of the light receiving area based on the binary image data, determination of a reference position, and inspection range based on the reference position And a step of inspecting a defect of the light receiving element based on the image data within the inspection range.In the step of setting the inspection range, the dead zone region existing between the first light receiving region and the second light receiving region facing each other included in the extracted light receiving regions can be removed. The image data of the first light receiving region and the second light receiving region are enlarged or reduced to create one apparent light receiving region, and the shape and shape of each light receiving region after the dead zone region is removed in advance. The shape of each light receiving area of the set setting pattern is compared, and the reference position is determined by the light receiving area having the most correlation.
[0020]
  Preferably,In the step of setting the inspection range, the inspection range setting unit apparently has the same total area of one light receiving region and the sum of the areas of the first light receiving region and the second light receiving region. As described above, the image data of the first light receiving area and the second light receiving area are enlarged or reduced.
[0021]
  Preferably, in the step of setting the inspection range, four circumscribed sides circumscribing each other at a predetermined angle circumscribing the light receiving area having the most correlation are calculated and formed by the four circumscribed sides. The center of gravity of the circumscribed rectangleAboveDetermine as the reference position.
[0023]
Preferably, in the step of performing the inspection, the quality of the light receiving element is determined based on an area in the inspection range of the shape of the light receiving region extracted based on the binary image data.
[0024]
Preferably, in the step of performing the inspection, the quality of the light receiving element is determined based on a density value corresponding to the intensity of the light reception signal for each scanning point within the inspection range of the image data.
[0025]
According to the inspection method for a light receiving element of the present invention, the light receiving area of the light receiving element is irradiated with a predetermined level of inspection light, and the inspection light is scanned on the light receiving area, and the light receiving element at each scanning point of the inspection light Image data is created based on the received light signal from the image, the image data is binarized, the shape of the received light area is extracted based on the binary image data, and a reference position serving as a reference for a predetermined inspection range And detecting the defect of the light receiving element based on the image data within the inspection range with the reference position as a reference.
By setting the inspection range as described above and inspecting the defect of the light receiving element from the image data within the inspection range, the inspection process of the light receiving element from the image data can be easily performed, and the inspection of the light receiving element Can be performed with high accuracy.
[0026]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of a light receiving element inspection apparatus and inspection method of the present invention will be described below with reference to the drawings.
[0027]
FIG. 1 is a schematic configuration diagram of a light receiving element inspection apparatus according to the present embodiment.
[0028]
  1 includes a laser light source 11, a beam expander 12, a galvano lens 13, an X direction actuator 14a, a Y direction actuator 14b, a scanner driver 15, a relay lens assembly 16, an objective lens 17, an object to be inspected. The workpiece 18 includes a workpiece 18, a socket 19, a control unit 21, a scanner controller 22, a conversion unit 23, a data processing unit 24, a display unit 25, and a power supply unit 26.
  The laser light source 11, the beam expander 12, the galvano lens 13, the X-direction actuator 14a, the Y-direction actuator 14b, and the scanner driver 15 are used for light irradiation in the claims.PartIt corresponds to.
[0029]
The laser light source 11 is made of, for example, a laser diode, and emits laser light having a wavelength actually used for a light receiving element that is an object to be inspected toward the beam expander 12, for example. For example, when inspecting a light receiving element used for a CD, a laser light source that outputs laser light having a wavelength of 780 nm and an output of 2 mW is used.
[0030]
The beam expander 12 converts a thin parallel light beam of the laser light 12 emitted from the laser light source 11 into a thick parallel light beam.
[0031]
The galvano lens 13 reflects the laser beam emitted through the beam expander 12 and guides the laser beam toward the workpiece 18.
[0032]
The X-direction actuator 14a moves the galvano lens 13 in the X-axis direction, thereby scanning the laser light reflected by the galvano lens 13 and guided toward the workpiece 18 in the X direction.
The Y-direction actuator 14b moves the galvano lens 13 in the Y-axis direction, thereby scanning the laser light reflected by the galvano lens 13 and guided toward the workpiece 18 in the Y direction.
[0033]
The scanner driver 15 receives a control signal from the scanner controller 22 and generates a drive voltage necessary for driving the X-direction actuator 14a and the Y-direction actuator 14b, and uses the drive voltage as the X-direction actuator 14a and the Y-direction actuator. To 14b.
[0034]
The relay lens assembly 16 irradiates the laser beam reflected by the galvano lens 13 in the direction of the workpiece 18. For example, the relay lens assembly 16 includes a θ lens 16a, a relay lens 16b, and an imaging lens 16c.
[0035]
The objective lens 17 focuses the laser beam guided from the relay lens assembly 16 on the light receiving element.
[0036]
As shown in FIG. 19A, the workpiece 18 includes a semiconductor chip 100 including a light receiving element 101 that is an object to be inspected. As described above, the workpiece 18 is formed on the semiconductor chip 100. A lead frame 103 connected to the output terminals of the respective light receiving regions (A to F) of the light receiving element 101 is formed.
[0037]
The socket 19 is mounted and fixed with a work 18 and is formed with electrode pads and the like for connecting to a lead frame and the like connected to the output terminal of each light receiving area of the light receiving element. An output signal from the element is output to the conversion unit 23.
[0038]
The control unit 21 performs various controls on the data processing unit 24 and the scanner controller 22.
For example, the control unit 21 sets a start point for scanning, a scan speed, and a scan size, and outputs a control signal corresponding to the settings to the scanner controller.
[0039]
The scanner controller 22 outputs a control signal to the scanner driver 15 based on a control signal from the control unit 21.
[0040]
As will be described in detail later, the conversion unit 23 performs a predetermined process on the output signal from the light receiving element input via the socket 19 and outputs the processed signal to the data processing unit 24.
[0041]
Although the details will be described later, the data processing unit 24 performs a predetermined process on the image data obtained by the conversion unit 23 to inspect the light receiving element.
[0042]
The display unit 25 displays image data, processing results, and the like obtained by the data processing unit 24 on the screen.
[0043]
The power supply unit 26 supplies power to the semiconductor chip of the work 18 through the socket 19.
[0044]
The operation of the inspection apparatus according to the present embodiment having the above configuration will be described.
Laser light emitted from the laser light source 11 is incident on the galvano lens 13 via the beam expander 12, reflected by the galvano lens 13, guided to the objective lens 17 via the relay lens assembly 16, and received by the light receiving element. It will be focused on the area.
Then, an electrical signal corresponding to the amount of incident light is output from the light receiving element that has received the laser light to the conversion unit 23, subjected to predetermined processing, and output to the data processing unit 24. In the data processing unit 24, a predetermined process is performed on the output signal to generate image data.
A drive voltage is generated by the scanner driver 15 that has received a control signal output from the scanner controller 22 based on a control signal from the control unit 21, and the X-direction actuator 14a and the Y-direction actuator that have received the drive voltage. 14b scans the galvano lens 13, whereby the laser beam is scanned in the X-Y direction in the light receiving region of the light receiving element, and image data is created in the data processing unit 24 from the output signal of the light receiving element, and is described later. Is performed.
[0045]
FIG. 2 shows an example of the configuration of the conversion unit 23.
As shown in FIG. 2, the conversion unit 23 includes a gain adjuster 31, an adder 32, an A / D converter 33, and an interface (I / F) unit 34.
It is assumed that the light receiving element 101 has rectangular light receiving areas A to F divided into six as shown in FIG. 19B, for example.
[0046]
When the electrical signals (SA to SF) generated by the light receiving regions (A to F) of the light receiving element 101 corresponding to the laser light are respectively input, the gain adjuster 31 performs gain adjustment of the signals and gain The adjusted signal is output to the adder 32. For example, in the gain adjuster 31, when the output of the light receiving element 101 is the rated output, the gain of each signal is adjusted so as to be 1V ± 0.2% at the subsequent input to the A / D converter 33.
[0047]
When the gain-adjusted signal (SA to SF) is input from the gain adjuster 31, the adder 32 adds the signals from the respective light receiving areas. This addition processing is later performed in the data processing unit 24 in order to perform image processing as output signals of the entire plurality of light receiving regions.
Here, for example, when the laser light is scanned from the light receiving region E to the light receiving region F and an electric signal is taken out, for example, when the laser light is irradiated to the light receiving region E, an output signal from another light receiving region Is substantially 0, even if signals from the respective light receiving areas (SA to SF) are added later by the adder 32, the signals in the respective light receiving areas do not mix and the light receiving areas corresponding to the scanning points. Can be detected as an output signal from.
The adder 32 corresponds to the calculation unit in the claims.
[0048]
The A / D converter 33 converts the analog signal added by the adder 32 into a digital signal for input to the digital input data processing unit 24. For example, the A / D converter 33 uses a 16-bit A / D converter in order to compensate for the absolute accuracy of subsequent image processing.
[0049]
The interface unit 34 performs an interface for outputting the digital signal input from the A / D converter 33 to the data processing unit 24. For example, when the data processing unit 24 can only process 8-bit data, if the 16-bit data from the A / D converter 33 is handled as it is, the resolution will be about 0.4%. Data less than half of the scale data is discarded and output to the data processing unit 24. Thereby, the resolution can be improved to about 0.2%.
[0050]
The operation of the conversion unit 23 will be described.
[0051]
When an electric signal from the light receiving area (A to F) at each position as shown in FIG. 19B is input to the gain adjuster 31, the gain of the signal is adjusted and the gain adjusted signal is added. To the device 32.
When the gain-adjusted signal (SA to SF) is input from the gain adjuster 31 to the adder 32, the signals from the respective light receiving regions are added, and as a result, A / It is output to the D converter 33.
The added analog signal from the adder 32 is input to the A / D converter 33, converted into a digital signal, and output to the data processing unit 24 via the interface unit 34. The light receiving element to be inspected will be inspected.
[0052]
  FIG. 3 shows an example of the configuration of the data processing unit 24.
  As shown in FIG. 3, the data processing unit 24 roughly includes an image data creation unit 40, an inspection range setting unit 50, and an inspection unit 60.
  The inspection range setting unit 50 includes a binarization processing unit 41, an enlargement / reduction unit 42, a shape extraction unit 43, a reference position determination unit 44, and a position correction unit 45.
  The inspection unit 60 includes an area calculation unit 46, a primary average value calculation unit 47, a secondary average value calculation unit 48, and a determination unit 49.
  Note that the data processing unit 24 performs the data processing in the claims.PartIt corresponds to.
[0053]
The image data creation unit 40 stores the digital data input via the interface unit 34 of the conversion unit 23 in, for example, a memory (not shown) corresponding to the scanning position based on the signal from the control unit 21, and the storage Image data is created based on the processed data.
[0054]
When the multivalued image data created by the image data creation unit 40 is input, the binarization processing unit 41 is based on the density value (lightness of each pixel) corresponding to the intensity of the received light signal in the image data. The image data is binarized.
Specifically, the density value of each pixel is compared with a predetermined threshold value, and when the density value is smaller than this threshold value, a value of “1” (dark) is given to the pixel, When the value is larger than this threshold value, a value of “0” (bright) is given to the pixel.
By applying binarization processing to the multivalued image data, the boundary between the light receiving area and the dead zone becomes clear, and the outline shape of the light receiving area is specified.
In the specification of the present application, the image data of the light receiving area formed by performing image processing with signals from each light receiving area is referred to as pad data.
[0055]
The enlargement / reduction unit 42 enlarges or reduces each pad data binarized and specified by the binarization processing unit 41.
Specifically, each pad data is expanded, the dead zone between pad data is filled, one pad data is created, and the size of pad data is reduced by the amount expanded to create the one pad data. To do.
This is to reduce the number of target pad data for determining the reference position, thereby shortening the measurement processing time. Also, by filling the dead zone, holes, cracks in the dead zone DZ outside the pad data, This is to remove noise components such as isolated points.
[0056]
The shape extraction unit 43 calculates, for example, a circumscribed rectangle of each pad data from the binary image data.
[0057]
The reference position determination unit 44 determines the reference position of the image data in the light receiving area.
When the reference position is set, binary image data is input from the binarization processing unit 41 and multi-value image data is input from the image data creation unit 40, and information on the reference position is input to the binary and multi-value image data. In addition, it is output to the position correction unit 45.
As a specific method of determining the reference position, for example, when the light receiving area is rectangular, the shape extraction unit 43 most correlates with the master pattern among the circumscribed rectangles of the individual pad data. A circumscribed rectangle having a large value is obtained, and the side centroid of the circumscribed rectangle having the maximum correlation value is determined as a reference position.
[0058]
The position correction unit 45 detects the inclination of a circumscribed rectangle in the binary image data to which the reference position information is added by the reference position determination unit 44, and the binary and multi-value images so that the inclination follows a predetermined reference direction. Transform data into coordinates.
That is, in the image data of the light receiving area that is only binarized, the contour shape of the light receiving area is specified, but the inclination of the light receiving area is not constant for each image data.
Therefore, as will be described later, in order to detect the area of the image data, the inclination of the image data in the light receiving region is detected, and the inclination of the light receiving region is centered on the reference position of the image data in a predetermined reference direction. The image data is rotated so as to follow.
The position correction unit 45 can perform the alignment between the binary image data and the mask pattern, and sets the area inspection range in the binary image data.
Further, the position correction unit 45 sets the inspection range excluding the peripheral portion in the above-described area inspection range as the density value inspection range in the multivalued image data.
[0059]
The area calculation unit 46 receives binary image data in which a predetermined inspection range is set by the inspection range setting unit 50, and calculates the area of each pad data of the binary image data in the inspection range.
[0060]
The primary average value calculation unit 47 receives multi-value image data in which a predetermined inspection range is set by the inspection range setting unit 24, and averages the density values of the pixels of the multi-value image data in the inspection range. A value (primary average value) is calculated, and an area of a pixel having a density having a predetermined deviation from the primary average value is calculated.
[0061]
The secondary average value calculation unit 49 further calculates an average value (secondary average value) excluding pixels having density values greater than or equal to a predetermined deviation from the primary average value, and from the secondary average value The area of a pixel having a predetermined deviation density is calculated.
[0062]
The determination unit 49 inputs each data calculated by the area calculation unit 46, the primary average value calculation unit 47, and the secondary average value calculation unit 48, and determines the quality of the light receiving element.
Specifically, it is evaluated whether or not the area calculated by the area calculation unit 46 satisfies a set value.
Further, it is determined whether or not the primary average value calculated by the primary average value calculation unit 47 has reached a set value, and the area of a pixel having a predetermined deviation density from the primary average value is predetermined. Determine whether it is within the set value.
Further, it is determined whether or not the area of a pixel having a density of a predetermined deviation from the secondary average value calculated by the secondary average value calculation unit 48 is within a predetermined set value.
In the determination of the area, the primary average value, and the secondary average value, if even one of the set values is not satisfied, the light receiving element is determined to be defective.
[0063]
Next, regarding the inspection method of the light receiving element by the inspection apparatus according to the above-described embodiment, the image of the light receiving region on the display screen 25a of the display unit 25 illustrated in the flowcharts of FIGS. Will be described with reference to FIG.
[0064]
First, as shown in FIG. 7A, the multivalued image data created by the image data creation unit 40 is input to the binarization processing unit 41 (step ST1), and is displayed on the display screen 25a of the display unit 25. The displayed pad data (PA to PF) of each light receiving area is binarized based on the brightness (density value) of each pixel (step ST2).
[0065]
Next, as shown in FIG. 7B, the enlargement / reduction unit 42 enlarges the pad data (PA to PD), fills the dead zone DZ between the pad data (PA to PD), and generates one pad data. PAD ′ is created (step ST3).
As described above, this is because the number of target pad data for determining the reference position later is reduced to shorten the measurement processing time.
Another reason is to remove noise components such as holes, cracks and isolated points in the dead zone DZ outside the pad data by filling the dead zone DZ.
[0066]
Next, as shown in FIG. 7C, after reducing the number of target pad data, the enlargement / reduction unit 42 further enlarges the pad data PAD ′ so as to create pad data PAD ′. Is reduced to create pad data PAD so that the total area of pad data PA to PD is equal to the area of pad data PAD (step ST4).
[0067]
Next, for example, a circumscribed rectangle of each pad data (PE, PAD, PF) is calculated from the binary image data by the shape extraction unit 43 (step ST5).
[0068]
Next, the reference position determination unit 44 compares the major axis and minor axis of the circumscribed rectangle of each pad data with the major axis and minor axis of the master pattern rectangle stored in the memory of the data processing unit 24 in advance. The circumscribed rectangle having the maximum correlation value is calculated (step ST6).
[0069]
Next, as shown in FIG. 8D, for example, when the circumscribed rectangle of the pad data PE has the maximum correlation value, it is the intersection of the center line of the vertical line and the horizontal line of the circumscribed rectangle of the pad data PE. The side gravity center G is obtained and determined as a reference position (step ST7).
The reason why the side centroid is adopted is that, in the case of the area centroid, the area centroid position is affected by a void or the like in the pad data.
[0070]
Next, the position correction unit 45 detects the inclination of the circumscribed rectangle having the maximum correlation value in the image data determined at the reference position G, and is binarized so that the inclination follows a predetermined reference direction. Is transformed (step ST8).
Specifically, for example, as shown in FIG. 8E, an equivalent ellipse C having the same area as the circumscribed rectangle of the pad data PE having the maximum correlation value is calculated, and the horizontal axis of the principal axis C1 of the equivalent ellipse C is calculated. The angle θ with respect to the axis (Y-axis) is calculated. When the angle θ is 90 °, the inclination is zero, and when the angle is 90 + α °, the inclination is α °.
In the figure, the horizontal axis (Y axis) and the vertical axis (X axis) are set in a predetermined direction in advance.
Then, if there is an inclination after calculating the angle θ, for example, affine transformation is applied to the image data so that the inclination of the light receiving area follows a predetermined reference direction centering on the reference position G of the image data. Rotate the image data.
[0071]
Next, as shown in FIG. 9F, the reference position (side centroid) G of the pad data PE having the maximum correlation value and the side centroid MG of the pad data ME of the corresponding master pattern (MA to MF). The master patterns (MA to MF) are moved so as to match.
[0072]
Next, as shown in FIG. 9G, the logical product of the binary image data (PA to PF) and the master pattern (MA to MF) whose positions are corrected by the area calculation unit 46 with the reference position G as the center. (AND processing) is taken (step ST9).
If the area of each pad data (PA to PF) in the binary image data has a predetermined set value in the area of the master pattern (MA to MF), there is no problem as a device. This is to eliminate the influence of voids and the like of binary image data outside the area of the master pattern (MA to MF).
[0073]
Next, each pad data (PA ′ to PF ′) in the binary image data after the logical product with the master pattern shown in FIG. 9 (g) is labeled (numbered for each pad data) (step ST10). Then, the area for each pad data (PA ′ to PF ′) after labeling is measured by the area calculation unit 46 (step ST11).
[0074]
Next, the determination unit 49 evaluates whether or not the specified value area is satisfied for each pad data (PA ′ to PF ′) after labeling (step ST12).
Then, only when all the pad data (PA ′ to PF ′) satisfy the specified area, the area evaluation is made good, and the process proceeds to the next inspection step, and the specified area is also set for one pad data (PA ′ to PF ′). If not satisfied, the subsequent inspection is stopped as a defective product (step ST13).
In addition, in the determination of the area, those in which the pad pattern does not have a predetermined coincidence with the master pattern due to the logical product of the binary image data (PA to PF) and the master pattern (MA to MF) are also regarded as defective products. Determined.
[0075]
Next, when it is determined in step ST12 that the specified value area is satisfied, the process proceeds to measurement of density values in the multi-value image data.
Specifically, as shown in FIG. 10, the density value in the examination range (RA to RF) of the density value excluding the peripheral corner portion of each pad data (PA to PF) by the primary average value calculation unit 47. The average value (primary average value) is calculated (step ST14). For example, the inspection range (RA to RF) is based on the visual standard.
[0076]
Next, the determination unit 49 determines whether or not the primary average value within the inspection range of each pad data (PA to PF) is within the set value range (step ST15).
If the primary average value is within the set value range, the process proceeds to the next inspection step as good. If the primary average value is not within the set value range, the subsequent inspection is performed as a defective product. Is canceled (step ST13).
[0077]
Next, when it is determined in step ST15 that the value of the primary average value is good, the primary average value calculation unit 47 further sets a density value of deviation −X% from the primary average value as a threshold value. Then, the multi-valued image data is binarized (step ST16), and the area of the pixel having the density value of deviation −X% is measured (step ST17).
FIG. 11 shows the density value for each position in a predetermined direction. In FIG. 11, the vertical axis indicates the density value, and the horizontal axis indicates the position.
As shown in FIG. 11, the multi-valued image data is binarized using a density value of deviation −X% from the primary average value M as a threshold value, and pixels having density values of deviation −X% or less (in the figure, Area) (shown in black area).
[0078]
Next, the determination unit 49 determines whether the area of the pixel having a density value of deviation −X% or less is within the visual standard (pixel corresponding to a diameter of 10 μm) (step ST18). Advances to the next inspection step, and if it is out of the visual standard, it is determined as a defective product, and the subsequent inspection is stopped (step ST13).
In addition, the evaluation of the area of the pixel having the density value of deviation + X% or more is not added to the above evaluation due to a void in the vicinity of the pad data (PA to PF), etc. This is because a void outside the pad data (including the vicinity) has no problem in use of the device.
[0079]
Next, when it is determined that the value is good in step ST18, the secondary average value calculation unit 48 removes pixels having a density value equal to or less than −X% from the primary average value, and determines from the primary average value. An average value (secondary average value) of density values of pixels having a density value of deviation −X% or more is calculated (step ST19).
This is because, for the primary average value, density values that are extremely deviated from the average value within the inspection range (low density) are also added to calculate the average value, so that the average value drops and an error of this amount occurs. Because there is a possibility.
Therefore, in order to increase the accuracy of the inspection, the average value (secondary average value) is calculated once again, excluding the extremely biased density value.
[0080]
Further, the secondary average value calculation unit 48 binarizes the multi-value image data using the density value of deviation −X% from the secondary average value as a threshold value (step ST20), and the deviation −X% or less. The area of the pixel having the density value is measured (step ST21).
[0081]
Next, the determination unit 49 determines whether the area of the pixel having a density value of deviation −X% or less is within the visual standard (pixel corresponding to a diameter of 10 μm) (step ST22). Is determined to be a non-defective product (step ST23), and if it is out of the visual standard, it is determined to be a defective product (step ST13).
Note that the evaluation of the area of the pixel having a density value of deviation + X% or more is not added to the above evaluation for the same reason as the evaluation using the primary average value.
[0082]
Next, FIG. 12 to FIG. 17 show image data of non-defective and defective light receiving elements, respectively.
[0083]
First, the pad data in each of the cases where the light receiving element is good or bad will be described with reference to FIGS. In FIGS. 12 to 14, in addition to the image data of the light receiving element, outputs x1 to x5 from the respective light receiving regions and YY ′ when the laser beam is scanned in the X direction along the line XX ′. Outputs y1 to y5 from the respective light receiving regions when the laser beam is scanned in the Y direction along the line are also shown. Further, each figure shows an average value of the density of all the light receiving regions.
[0084]
FIGS. 12A and 12B show images of light receiving elements determined as non-defective products by the inspection method using the inspection apparatus according to this embodiment, and the images shown in FIGS. 12A and 12B are shown. In particular, there is no influence based on the peeling of the transparent resin covering the void or the light receiving element, the distribution of the output waveforms (x1, y1) and (x2, y2) is not particularly varied, and the average density is also It was confirmed that the value was within the specified range.
[0085]
FIGS. 13A and 13B show images of light receiving elements determined as defective by the inspection method using the inspection apparatus according to the present embodiment, and are shown in FIGS. 13A and 13B. In the image, there are no particular effects due to voids or peeling of the transparent resin covering the light receiving element, and there is no particular variation in the distribution of the output waveforms (x3, y3) and (x4, y4). Since the average values of the light receiving elements are as small as 67 and 68 and are not within the specified value range, the light receiving element is determined to be defective in step ST15.
[0086]
In the image from the output of the light receiving element shown in FIG. 14, there is no particular influence due to voids or peeling of the transparent resin covering the light receiving element, but there is a variation in the distribution of the output waveform x5 in particular, and the average density Therefore, in step ST15 (also in step ST18), the light receiving element is determined to be defective.
[0087]
Next, determination of pass / fail of the light receiving element will be described using the image data shown in FIGS.
[0088]
In the image from the output of the light receiving element shown in FIG. 15, a striped pattern that cannot be confirmed in the image from the non-defective light receiving element shown in FIG. This has an effect due to the peeling of the transparent resin covering the light receiving element, and in step ST18, the light receiving element was determined to be defective.
[0089]
In the image from the output of the light receiving element shown in FIG. 16, voids scattered in the mottle can be clearly confirmed. This is due to the influence of voids existing on the surface or inside of the light receiving element. Therefore, in step ST18, the light receiving element is determined to be defective.
[0090]
In the image from the output of the light receiving element shown in FIG. 17, it is clearly confirmed that there is a shadow in the upper left part in the figure. This is considered to be due to the influence of foreign matter present in the transparent resin covering the light receiving element. Therefore, in step ST18, the light receiving element was determined to be defective.
[0091]
As described above, according to the above-described light receiving element inspection apparatus according to this embodiment, a laser having a wavelength that is actually used is not inspected for the presence or absence of defects in the light receiving element by imaging the light receiving element or the like. By scanning the light in the light receiving area of the light receiving element and performing image processing from the output obtained from the light receiving element, for example, peeling of the transparent resin covering the light receiving element, foreign matter in the transparent resin, dirt, Since various defects such as a decrease in sensitivity affect the image data, it is possible to accurately determine whether the light receiving element is good or bad by selecting the level of the influence from the defect.
Therefore, even when there is a difference in sensitivity between the light receiving areas, which cannot be handled by the conventional method, it can be determined as defective.
In addition, there is no problem of inspection level variations in human visual inspection and inspection level variations in illumination conditions in image processing inspection as in the prior art, and stable inspection can be performed.
[0092]
Further, when inspecting by image processing, a circumscribed rectangle of each pad data is calculated, a reference position is determined, position correction of each pad data is performed based on the reference position, and an inspection range is determined based on the reference position. Set and evaluate the area and density value in the inspection range, so it is not necessary to calculate coordinates etc. like pattern matching used in normal image processing, so processing speed is increased Can be achieved.
Further, when determining the reference position, the processing speed can be further increased by enlarging and reducing the pad data of the complex light receiving area to reduce the number of target shapes.
Furthermore, by performing the position correction around the center of gravity of the side, it is possible to easily calculate a reference position that is not affected by a void or the like in the pad data and set an appropriate inspection range.
[0093]
Embodiments of the light receiving element inspection apparatus and inspection method of the present invention are not limited to the above description.
For example, in the present embodiment, the inspection of the light receiving element for CD as the light receiving element has been described.
For example, in the labeling process of the image data from the light receiving element for MD, for example, as shown by the arrow in FIG. 18, the numbering is not performed in the order of pad data PI, PD, PA, PJ. There is a case. For example, as shown in FIG. 18, when the pad data PD is shifted to the left in the figure with respect to other pad data (PI, PA, PJ), the appearance order of the pad data PD detected first is 1 It will be a turn. For example, in order to avoid this, various modifications such as rearranging by the Y coordinate of each pad data after numbering can be performed for each light receiving element.
[0094]
For example, in the present embodiment, the relay lens assembly 16 including the θ lens 16a, the relay lens 16b, and the imaging lens 16c, the objective lens 17, and the like are used as optical members that guide the laser light to the light receiving element. The present invention is not limited to this as long as the laser light can be focused on the light receiving region of the light receiving element.
Further, in the present embodiment, the galvano lens 13 is scanned by the X direction actuator 14a and the Y direction actuator 14b to scan the laser light in the light receiving region of the light receiving element. However, the laser light is received by the light receiving element. Other forms are possible as long as the region can be scanned.
In addition, various modifications can be made without departing from the scope of the present invention.
[0095]
【The invention's effect】
ADVANTAGE OF THE INVENTION According to this invention, the precision and stability of a test | inspection of a light receiving element can be improved, and a reliable light receiving element can be provided as a good product.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a light receiving element inspection apparatus according to the present invention.
FIG. 2 is a diagram showing in detail the configuration of a conversion unit shown in FIG. 1;
FIG. 3 is a diagram illustrating an example of a configuration of a data processing unit illustrated in FIG. 1;
FIG. 4 is a flowchart of processing in the light-receiving element inspection method according to the present embodiment.
FIG. 5 shows a continuation of the flowchart shown in FIG.
FIG. 6 is a continuation of the flowchart shown in FIG. 5;
FIG. 7 shows an image of the light receiving element on the display screen of the display unit in each process of the light receiving element inspection method according to the present embodiment.
8 shows an image of the light receiving element on the display screen of the display unit in the processing subsequent to FIG.
9 shows an image of the light receiving element on the display screen of the display unit in the continuation process of FIG. 8. FIG.
FIG. 10 shows an image of a light receiving element on the display screen of the display unit in the continuation process of FIG. 9;
FIG. 11 shows density values for each position in a predetermined direction.
FIG. 12 shows an image from the output of a light receiving element determined to be non-defective.
FIG. 13 shows an image from the output of a light receiving element that is determined to be defective because the average density is smaller than a specified value.
FIG. 14 shows an image from the output of a light receiving element that is determined to be defective because the average density is smaller than a specified value.
FIG. 15 shows an image from the output of the light receiving element when the transparent resin covering the light receiving element is peeled off.
FIG. 16 shows an image from the output of the light receiving element when voids exist on the surface or inside of the light receiving element.
FIG. 17 shows an image from the output of the light receiving element when there is a foreign substance in the transparent resin covering the light receiving element.
FIG. 18 shows one mode of modification of labeling processing when labeling each pad data in image data from a light receiving element for MD.
FIG. 19A is a cross-sectional view of a semiconductor chip having a light-receiving element used for, for example, a CD, and FIG. 19B is a light-receiving formed on the semiconductor chip of FIG. 19A. It is an enlarged plan view of an element.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Inspection apparatus, 11 ... Laser light source, 12 ... Beam expander, 13 ... Galvano lens, 14a ... X direction actuator, 14b ... Y direction actuator, 15 ... Scanner driver, 16 ... Relay lens assembly, 16a ... θ lens, 16b DESCRIPTION OF SYMBOLS ... Relay lens, 16c ... Imaging lens, 17 ... Objective lens, 18 ... Workpiece, 19 ... Socket, 21 ... Control part, 22 ... Scanner controller, 23 ... Conversion part, 24 ... Data processing part, 25 ... Display part, 26 DESCRIPTION OF SYMBOLS ... Power supply part 31 ... Gain adjuster 32 ... Adder 33 ... A / D converter 34 ... Interface part 40 ... Image data creation part 41 ... Binarization processing part 42 ... Enlarging / reducing part, 43 ... Shape extraction unit, 44 ... Reference position determination unit, 45 ... Position correction unit, 46 ... Area calculation unit, 47 ... Primary average value calculation unit, 48 ... Secondary average value calculation , 49 ... determination part, 50 ... inspection range setting part, 60 ... inspection part, 101 ... light receiving element, 102 ... transparent resin, 103 ... lead frame, 104 ... wire, A, B, C, D, E, F ... Light receiving area.

Claims (18)

In order to inspect a light receiving element that includes a plurality of light receiving regions and outputs a light receiving signal according to the light irradiation level to the light receiving region,
A light irradiation unit that irradiates a light receiving region of the light receiving element with a predetermined level of inspection light and scans the light receiving region over the light receiving region;
A data processing unit including a processing system that creates image data and performs predetermined processing based on a light reception signal from the light receiving element at each scanning point of the inspection light, and
The processing system of the data processing unit is:
An inspection range setting unit that binarizes the image data, extracts a shape of the light receiving region based on the binary image data, determines a reference position, and sets an inspection range based on the reference position;
Based on the image data within the inspection range set in the inspection range setting unit, possess an inspection unit for inspecting defects of the light receiving element,
The inspection range setting unit
The first light-receiving region and the second light-receiving region included in the extracted light-receiving regions can be removed so as to remove the dead zone region existing between the first light-receiving region and the second light-receiving region facing each other. Each image data in the light receiving area is enlarged or reduced to create one apparent light receiving area, and after removing the dead zone, the shape of each light receiving area and the shape of each light receiving area of a preset setting pattern , And the reference position is determined by the light receiving area having the most correlation . The inspection range setting unit includes the first light receiving region so that the apparent area of the entire light receiving region is equal to the sum of the areas of the first light receiving region and the second light receiving region. 2. The light receiving element inspection apparatus according to claim 1, wherein image data of the region and the second light receiving region is enlarged or reduced . The inspection range setting unit may calculate the light-receiving region 4 enclosing sides possible cross each other at a predetermined angle circumscribes respect with the most correlation, the edges centroid of the circumscribing rectangle formed by the four circumscribing sides inspection apparatus of the light-receiving element according to claim 1 or 2, wherein determining as a reference position. The light-receiving element inspection device according to claim 1, wherein the inspection unit determines whether the light-receiving element is good or not based on an area within the inspection range of the shape of the light-receiving area extracted based on the binary image data. The light-receiving element inspection device according to claim 1, wherein the inspection unit determines the quality of the light-receiving element based on a density value corresponding to the intensity of the light-receiving signal for each scanning point within the inspection range of the image data. The light receiving element inspection apparatus according to claim 5 , wherein the inspection unit calculates an average value of the density values in the inspection range of the image data, and determines whether the light receiving element is good or bad using the average value. The light receiving element inspection apparatus according to claim 6 , wherein the inspection unit determines whether the light receiving element is good or bad by comparing the average value with a predetermined set value. The light receiving element inspection apparatus according to claim 6 , wherein the inspection unit determines the quality of the light receiving element based on a deviation of the density value within the inspection range from the average value. It further has a calculation unit that calculates a light reception signal corresponding to each light reception region,
The light receiving element inspection apparatus according to claim 1, wherein the data processing unit creates the image data of the light receiving element from the light receiving signal calculated by the calculating unit. In order to inspect a light receiving element that includes a plurality of light receiving regions and outputs a light receiving signal according to the light irradiation level to the light receiving region,
Irradiating a light receiving region of the light receiving element with a predetermined level of inspection light, and scanning the inspection light on the light receiving region;
Creating image data based on a light reception signal from the light receiving element at each scanning point of the inspection light;
Binarizing the image data, extracting a shape of the light receiving area based on the binary image data, determining a reference position, and setting an inspection range based on the reference position;
Based on the image data within the inspection range, it has a step of performing an inspection of defects of the light receiving element,
In the step of setting the inspection range,
The first light-receiving region and the second light-receiving region included in the extracted light-receiving regions can be removed so as to remove the dead zone region existing between the first light-receiving region and the second light-receiving region facing each other. Enlarge or reduce each image data of the light receiving area, apparently create one light receiving area,
Compare the shape of each light receiving region after removing the dead zone region and the shape of each light receiving region of a preset setting pattern,
An inspection method for a light receiving element, wherein the reference position is determined by a light receiving region having the most correlation . In the step of setting the inspection range, the first area of the first light receiving region is apparently equal to the sum of the areas of the first light receiving region and the second light receiving region. The light receiving element inspection method according to claim 10, wherein the image data of the light receiving area and the second light receiving area are enlarged or reduced . In the step of setting the inspection range, four circumscribing sides circumscribing each other at a predetermined angle circumscribing the light receiving area having the most correlation are calculated, and a side centroid of a circumscribed rectangle formed by the four circumscribing sides The light receiving element inspection method according to claim 10 , wherein: is determined as the reference position. The method for inspecting a light receiving element according to claim 10, wherein in the step of inspecting, the quality of the light receiving element is determined based on an area of the shape of the light receiving region extracted based on the binary image data in the inspection range. The method for inspecting a light receiving element according to claim 10, wherein in the step of inspecting, the quality of the light receiving element is determined based on a density value corresponding to the intensity of a light receiving signal for each scanning point within the inspection range of the image data. . The light receiving element inspection method according to claim 14, wherein in the step of performing the inspection, an average value of the density values in the inspection range of the image data is calculated, and the quality of the light receiving element is determined using the average value. . The light receiving element inspection method according to claim 15 , wherein in the inspection step, the quality of the light receiving element is determined by comparing the average value with a predetermined set value. The light receiving element inspection method according to claim 15, wherein, in the step of performing the inspection, the quality of the light receiving element is determined based on a deviation of the density value within the inspection range from the average value. In the step of creating a pre-Symbol image data, it calculates a light reception signal corresponding to each light receiving region, the inspection method of the light-receiving element according to claim 10, wherein generating the image data of the light receiving element from the computed received signal.

Images