DE19927991A1 - Cell-based image sensor, combines signals from two photosensitive elements, receiving overlapping first and second image sections via two lenses, and outputs signal representing image - Google Patents

Abstract

The image sensor has first and second photosensitive elements (32) that react to light to produce first and second signals. There are first and second lenses (30) that focus light from overlapping first and second image sections onto the first and second photosensitive elements (32). A combining device (36) is provided to combine the signals from the two elements, and output a signal representing the image. An Independent claim is also included for an image detection method.

Description

The present invention relates generally to Image sensors and in particular on scanning devices (scan ner) for document scanners, copiers, facsimile and multi functional peripherals that include sensing elements.

In the past, different types of images were used sensors in scanners, copiers and facsimile devices (fax) used to take a picture line by line.

For example, in a conventional flatbed scanner an optical single lens system is used. Originally was an image on a document through a single lens focused that covers the entire width of the document and the image into a single photosensitive semiconductor chip supplies. The lens usually consists of several optical surfaces. The difficulty with this solution approach was that the gaps between the document and the lens and between the lens and the semiconductor chip had to be quite large, which meant that the scanner tended to be quite large or fat. This led to extremely unwieldy devices.

Optical single lens systems in flatbed scanners have very high "f" numbers. The "f" refers to the win under which light is collected. The higher the "f" number the less light is collected. The high "f" number The main reason is that in individual lens scans Very bright light sources are required.

A significant improvement has been seen in the generation of Flatbed scanners generated, where the optical path with Mirror is folded to reduce the height of the device. Unfortunately, this increased the cost of the device.  

First was a number of additional precision mirrors required. Second, the mirrors contributed to additional Loss of light at. Third was an expensive semiconductor chip required. The chip had to be long enough to have enough pixels to see the full width of the image. The The width of the chip usually had to be wider than the minimum to be an acceptable aspect ratio (length / width) to show. If the chip has too large an aspect ratio he would not be stable and could be during half conductor processes, for example a saw and a chip attachment, break. Because the chip for this circuit order wider than minimum required to achieve an acceptable Obtaining aspect ratio is more expensive than if the chip is made with a minimal circuit width could be. And fourth, the devices continued to tilt to be relatively thick.

To develop less expensive lenses and more compact devices, were SELFOC lenses in a different type of one Scanner sensor used, called the contact image sensor becomes. The SELFOC lenses make up a plurality of rods that are placed upright. The light path is along the axis of the bars. The material from which a SELFOC lens is made is produced, varies with respect to the refractive index the radial distance from the axis of the rod. The radial one Gradient of the refractive index allows the rod to move behaves like a lens. Each staff behaves like one Lens with a small field of view. By arranging the Bars next to each other in a line, they can be a linear one Transfer image. When the SELFOC lenses are in line the fields of view of each Stablin overlap se. To create a linear image, the bars must be on be cut a length and positioned in such a way that the image of each lens is a positive (non-inverted tiert) image with a 1: 1 gain.

Small image sensors are much more compact than single lenses systems. They are generally less expensive, have one  much smaller field depth and produce a smaller one Picture quality.

There are two ways to make lenses. One is using glass rods. These glass rods are made by immersing them in a doping material and through Allow the dopant to diffuse into the rods posed. It is the dopant that has the refractive index gradient supplies. Then the rods are cut to length cut, tied with an epoxy resin and the ends ge grind. This process tends to be relatively time consuming to be agile and expensive.

A cheaper approach is to ver plastic turn that extru with a number of extrusion processes each with a different refractive index dex. The multiple extrusions are melted later, to a more continuous gradient of the refractive index produce. The plastic SELFOCs generally always show another tiered index that does not change evenly changes so that they have poor image quality. A major disadvantage with conventional contact formation sor is that the photosensitive semiconductor chips cannot be made as a single piece nen. Therefore, a large number of chips with excessive pre Precise tolerances are cut and then excessive to be visually joined together to make the long Generate line that is required to make an adjustment to deliver the width of the document. This generation A long line of chips is the most difficult combination construction process in the manufacture of a contact image sensor. The Sensor chips are also the most expensive part of these sensors.

The semiconductor chips must be put together to form a to maintain continuous pixel spacing, even between the pixels between neighboring chips. It may over all 20 or a comparable number of semiconductors chips that are needed to make a contact image sensor  deliver, no accumulation of tolerance variations. The pixels of a conventional contact image sensor point a bad "fill factor". The pixels at the ends of the chip must be sized around the space between the chip and the minimum spacing for an active Area of the chip to be reduced to the saw edge ver regardless of the amount for the Chipsaw process is required. So that all pixels in the All of the array must respond equally to light Resize pixels just like the end pixel. The fill factor is the active area of the pixel di validated by the area ver for an entire pixel is available (1 / the pixel-per-inch specification). A 600- Pixel-per-inch contact image sensor typically has one Fill factor of 30%. A low fill factor means that the pixel collects less light, either the Light source can be made brighter or the scanner slower must be done. Because the technology at 1200 pixels per Inch approaching, it will not be possible to use contact image sensors to manufacture.

In a conventional contact image sensor are chip saw and Chip mounting processes in manufacturing difficult because the tolerance that the chip must have in order to be sawn which is very precise (10 µm is typical) and the achievement of merging tolerances to maintain the spacing hold is very difficult. It is too difficult to do this with standard automatic chip attachment devices respectively.

The sawing and fastening processes often cause the Semiconductor chip splinters and breaks, which is a minor Yield and high costs. The width of the chip is often about 0.686 mm (0.027 inches), with a length of 12.7 to 20.3 mm (0.5 to 0.8 inches). The non-square side ratio makes it very difficult to saw the chips. Well Of course, the fewer chips need to be brought together, the less chips are required. The aspect ratio,  which is acceptable, seems to be between 20 and 30 to 1, which means that manufacturers often make the chips wider, so that they can use fewer chips for the width, but this results in wasted chip area, which additionally costs.

The alignment is so complex that one tends to use it in general to perform manually. But this often leads to severe losses, almost up to some loss of earnings from 30%, even in mature operations. Misalignment often have to retire an entire array of a sensor chips result. Because the chips cost the most in one Represent contact image sensor, this loss of yield has a big impact on costs.

In the past, scanners had white light lighting and that Color separation was achieved using prisms or filters. The white light sources were typically a tube lamp, such as a fluorescent lamp. The lamps were not color stable and required a regulation to the In control intensity and color. They also had one limited lifetime.

With LEDs it is now possible to separate the colors through the To achieve lighting. Using three or more colors, the detector is read three times, with the Color of lighting for each lighting is changed. More than three LEDs can be used to realisti delivering scissor colors (i.e. verifying color rendering improve).

Another difficulty relates to mounting of the chips on PC boards. The structure of PC boards tends to swim to the edges of the board what means that from an optical point of view, is difficult the semiconductor chips in such a way to position it precisely with the optics of the sensor are aligned.  

A less expensive, compact system that is similar or similar has better quality than the single lens system long sought by professionals, but has long been the ex experts withdrawn in this area.

The object of the present invention is kos inexpensive, compact image sensor devices with one improved quality and image acquisition creating procedures.

This task is accomplished by an image sensor device Claims 1 and 9 and a method for detecting Images solved according to claim 17.

A set of plastic molded lenses is used to make one Line drawing on a set of photosensitive chips the. Each lens works with a chip to make a cell produce. The cells have slightly overlapping fields of view to allow tolerance for manufacturing. The Overlap can be more than the minimum in order to be new construction of the picture to help. The lenses can ver Have enlargements (reductions) such that the Chip can be smaller than the picture.

The lens in each cell can have a single optical ele ment, but mostly it will have multiple optical surfaces Chen have. The use of multiple optical waiters surfaces can be used to correct optical aberrations (Aberrations), such as color mapping mistake. The lenses can also be reflective or can be a mix of refractive and reflective Be lenses. The optical surface shapes can be spherical shaped, toric, bending, or a mixture of these.

Preferred embodiments of the present invention  are appended below with reference to the Drawings explained in more detail. Show it:

Fig. 1 is a schematic side view of a system containing the cell-based image sensor of the present invention;

Fig. 2 is a schematic representation of the light detection section of the present invention;

Fig. 3 (PRIOR ART) a section of the structure of the contact image sensor semiconductor chip array; and

Fig. 4 shows a partial section of the semiconductor chip assembly of the present invention.

Referring now to Fig. 1, there is shown an image capture device 10 which may be like a scanner, facsimile (fax), copier, etc. The imager 10 is set to scan an image 14 on the top surface of a document 12 which is to move longitudinally in the directions indicated by arrow 16 so that "a" scan line across the widths of the Document 12 can be detected at once.

Illumination to the top of document 12 is provided by a series of differently colored light emitting diodes (LEDs). LEDs 18 , 20 , 22 and 24 can be red, blue, green and yellow, not necessarily in that order to provide the ability to capture different color images 14 . The LEDs in Fig. 1 are shown in the plane of the page. The LEDs may also be in a single line perpendicular to the plane of FIG. 1. The lighting device may also be an end-lighting device.

During a scan, the light from one of the light emitting diodes 18 travels along path 26 . The light travels along line 28 through a molded plastic lens array 30 to the photosensitive semiconductor chip 32 which is embedded in a molded lead frame 34 .

For a full color picture, three scans used, generally as red, green and blue Scanning operations are called. The spectral distribution the ability of the scanner can be affected by each scanning process act to reproduce a real color. It can do more than one color LED per scan to be on to produce better spectral distribution.

The photosensitive semiconductor chips 32 are connected to a circuit arrangement 36 in order to receive signals therefrom and to convert them into signals which can be used by the special scanner, fax machine or copier, etc., ver. The cable 38 supplies the image capture device 10 with power.

Referring now to FIG. 2, the view width of an image sensor 40 of an image capture device 10 is shown (which is shown lengthwise in FIG. 1). The image sensor 40 consists of the shaped lens array 30 , which consists of individual lenses, for example 30 A, 30 B, 30 C, 30 D and 30 E. The lenses are shown schematically as single element lenses, but mostly they will be multi elements. As shown in FIG. 2, each of the lenses covers a cellular field of view, which is respectively for the lenses 30 A, 30 B, 30 C, 30 D and 30 E as a cellular field of view 42 A, 42 B, 42 C, 42 D and 42 E is designated. As can be seen, the fields of view overlap as overlaps 44 A, 44 B, 44 C and 44 D. The lens array 30 focuses the respective cellular fields of view 42 A, 42 B, 42 C, 42 D and 42 E on the respective photosensitive Semiconductor chips 32 A, 32 B, 32 C, 32 D and 32 E. Each lens and the associated chip form a cell of the image sensor 40 .

The molded lead frame 34 has registration features (not shown) that enable precise positioning of the optics 30 to the molded lead frame. The registration features allow the molded lead frame to be positioned accurately in the build equipment, and the registration features allow the finished scanner assembly to be precisely placed throughout the scanner assembly by the final manufacture. The precision registration enables the semiconductor chip 32 to be precisely placed relative to the molded lead frame and therefore to the optics 30 .

Known photosensitive semiconductor chips 50 A and 50 B are shown in FIG. 3 (PRIOR ART), to which reference is now made. A series of pixels 52 , 54 , 56 , 58 and 60 are formed in the surface of the semiconductor chip 50A. Similarly, the semiconductor chip 50B has pixels 62 , 64 , 66 , 68 and 70 . The distances 53 and 57 between pixels must be kept the same and must also be maintained between the pixels at the ends of the semiconductor chip, such as the distance 61 . There must be no tolerance accumulation over all of the ten or more semiconductor chips 50 . This means that the semiconductor chips 50 A and 50 B must be sawn with very tight tolerances such that the interspace, referred to as the space 72 , must be sawn with a tolerance of typically 5.08 µm (0.0002 inches).

The prior art pixels have a low "fill factor". That is, they must be reduced in size to the maximum possible in order to keep the spacing between all the pixels the same, including the pixels at the ends of the chips 60 and 62 . Putting the chip together to maintain a constant pixel spacing across the space between the chips is difficult. It is too difficult to do this with standard automatic die attach devices. The precision chip attachment can be done with very expensive custom equipment, but is usually done manually.

The sawing process for the semiconductor chips normally produces splintered and broken edges. The chips for conventional che contact image sensors cannot egg the normal amount Tolerate splintering and breaking, and the sawing process must be maintained at a fine level.

The assembly process can also chip the chip and break. The chips can be broken and broken by being hit against each other. This he manual or automatic chip mounting is required operation is more difficult than usual, with the failures are the reason for high yield losses.

In Fig. 4, to which reference is now made, semiconductor chips 32 A and 32 B are shown. It should be noted that the new semiconductor chips are approximately one third or less the size of the 50 A and 50 B semiconductor chips according to the prior art. Because the pixels have a fill factor of 100%, they are contiguous and there is no problem with pixel spacing. For example, the pixels of the semiconductor chip 32 A, which are the pixels 82 , 84 , 86 , 88 and 90 , can be positioned using conventional semiconductor manufacturing processes, since the exact position at which the image 14 falls on the semiconductor chip 32 A is not critical and corrected by image processing who will. The overlap 44 A and image processing means that the distance between the chips 104 is not critical. The distance from the sawn edge of the chip and the pixels 102 is also not critical.

In operation, document 12 with image 14 runs past image capture device 10 . A narrow band of image 14 is sequentially illuminated by light emitting diodes 18 , 20 , 22 and 24 , the light following the path indicated by arrow 28 to lens array 30 .

If the lens array is designed to be achromatic, the lens array can focus each of the illumination colors from the LEDs 18 , 20 , 22 and 24 onto the plane of the semiconductor chip 32 .

As shown in Fig. 2, the length of the image 14 is reflected in the lenses 30 A, 30 B, 30 C, 30 D and 30 E, with each of the portions of the image 42 A, 42 B, 42 C, 42 D and 42 E is typically focused with a reduction between 3: 1 and 6: 1 by the lens array 30 , the overlap being indicated by the overlapping regions 44 A, 44 B, 44 C and 44 D.

Each of the portions of the image 14 is focused on a photosensitive pixel on a semiconductor chip 32 . In the area of the overlap 44 , the object image 14 can also be imaged on the semiconductor chips 32 . It should be noted that the precise alignment is not critical since the image processing will adjust and remove the overlap to recreate the original image 14 .

It has been determined that the entire image capture device 10 can be made much cheaper while delivering higher quality than in the prior art.

For example, the molded lens array 30 is more than six times less expensive than a glass SELFOC array and more than twice less expensive than a plastic SELFOC array.

Furthermore, molded lead frames are by a factor of two, or more, less expensive than printed circuit boards that are required in the prior art. Fer ner shaped lead frame simplify the Together add further components. In addition, the half conductor chip due to the reduction between 3: 1 to 6: 1 in size to a third to a sixth of a be known contact image sensor can be reduced, which is a proportional reduction in costs. This  Order would also cover most of the cost of the 30% loss of building yield from known contact picture eliminate sensors.

Furthermore, the semiconductor chip alignment is no longer critical and will accelerate assembly. At the same time, the quality of the image will be better because the lens array can operate at a higher resolution than the SELFOC array in contact image sensors. Better resolution means that the image sensor 40 can operate at 1200 dots per inch or more, which is much more than the known system, which is limited to approximately 600 dots per inch.

SELFOC-based image sensors are limited to approximately 1 mm field depth when used in scanners at 600 pixels per inch. The new element 40 can be designed using conventional optical design techniques to have a much greater depth of field. This means that the distance between the document 12 and the lens array 30 is not as sensitive and can have a wider range of variation.

While the high f number is more light than a similar one SELFOC lens requires collecting the improvement of the Effective light. Because contact image sensors are fast Lose fill factor when the resolution is 600 pixels per inch exceeds the advantage of the full fill factor new sensor a superior light collection at higher Deliver pixel-per-inch resolutions. This is a drastic one Effect when the resolution increases.

The semiconductor chips 32 can use CCD or photodiode technologies. The chips can be made in CMOS, bipolar or amorphous silicon.

Claims (20)

1. Image sensor device which has the following features:
a first photosensitive element ( 32 A- 32 E) responsive to light to provide a first signal;
a second photosensitive element ( 32 A- 32 E) responsive to light to provide a second signal;
a first lens ( 30 A- 30 E) which focuses light from a first portion ( 42 A- 42 E) of the image ( 14 ) onto the first photosensitive element ( 32 A- 32 E);
a second lens ( 30 A- 30 E) that focuses light from a second portion ( 42 A- 42 E) of the image onto the second photosensitive element, the second portion of the image ( 14 ) being a portion of the first portion of the image ( 14 ) overlaps; and
combining means ( 36 ) for combining the first and second signals from the first and second photosensitive members ( 32 A- 32 E) and outputting a signal representing the image ( 14 ). 2. Sensor device according to claim 1, wherein the first and second photosensitive member apart stands. 3. Sensor device according to claim 1, wherein
the first and second photosensitive elements include a plurality of pixel elements ( 82-100 ), and
focuses the light from the first and second portions ( 42 A- 42 E) of the image onto less than the majority of the pixel elements. 4. Sensor device according to one of claims 1 to 3, wherein the first and the second lens element ( 30 A- 30 E) are a single unit. 5. Sensor device according to one of claims 1 to 4, wherein the first and second photosensitive element ( 32 A- 32 E) are on separate carriers. 6. Sensor device according to one of claims 1 to 4, wherein the first and the second photosensitive element ( 32 A- 32 E) are on a single carrier ( 34 ). 7. Sensor device according to one of claims 1 to 6, in which
the first and second lens elements ( 30 A- 30 E) invert the first and second portions ( 42 A- 42 E) of the image ( 14 ), and
the combining means ( 36 ) comprises means for reverting the image ( 14 ). 8. Sensor device according to one of claims 1 to 7, further comprising a light source ( 18 , 20 , 22 , 24 ) in the vicinity of the first and second lens elements ( 30 A- 30 E) for illuminating the image ( 14 ). 9. Image sensor device, which has the following features:
a first photosensitive semiconductor chip ( 32 A- 32 E) responsive to light to provide a first signal;
a second photosensitive semiconductor chip ( 32 A- 32 E) responsive to light to provide a second signal;
a first lens ( 30 A- 30 E) that focuses light from a first portion ( 42 A- 42 E) of the image ( 14 ) onto the first photosensitive semiconductor chip ( 32 A- 32 E); and
a second lens, the light from a second Ab-section (42 A-32 E) of the image on the second photosensitive semiconductor chip is focused, wherein the second portion of the image (14) overlaps a portion of the first portion of the image (14). 10. Sensor device according to claim 9, wherein the first and the second photosensitive semiconductor chip ( 32 A- 32 E) are spaced apart. 11. Sensor device according to claim 9 or 10, wherein the first and the second photosensitive semiconductor chip ( 32 A- 32 E) contain a plurality of pixel elements ( 82-100 ), and
the light from the first and second portions ( 42 A- 42 E) of the image ( 14 ) is focused on less than the majority of the pixel elements. 12. Sensor device according to one of claims 9 to 11, wherein the first and the second lens ( 30 A- 30 E) are a single unit which is formed from plastic. 13. Sensor device according to one of claims 9 to 12, wherein the first and the second photosensitive semiconductor chip ( 32 A- 32 E) are on separate chip carriers. 14. Sensor device according to one of claims 9 to 12, wherein the first and the second photosensitive semiconductor chip ( 32 A- 32 E) on a molded unique plastic carrier ( 34 ) capable of the first and second photosensitive semiconductor chip ( 32 A- 32 E) precisely with respect to the first and second lens ( 30 A- 30 E). 15. Sensor device according to one of claims 9 to 14, wherein the first and the second lens ( 30 A- 30 E) invert the first and the second section ( 42 A- 42 E) of the image ( 14 ). 16. Sensor device according to one of claims 9 to 15, which a plurality of light emitting diodes ( 18 , 20 , 22 , 24 ) in the vicinity of the first and second lens ( 30 A- 30 E) to illuminate the image ( 14 ) , having. 17. A method for capturing images ( 14 ), comprising the following steps:
Focusing light from a first portion ( 42 A- 42 E) of the image ( 14 ) onto a first photosensitive element ( 32 A- 32 E);
Focusing light from a second portion ( 42 A- 42 E) of the image ( 14 ) onto a second photosensitive element ( 32 A- 32 E), the second portion of the image ( 14 ) overlapping a portion of the first portion of the image ;
Combining the first and second signals from the first and second photosensitive elements ( 32 A- 32 E). 18. The method according to claim 17, wherein
the step of focusing light from a first portion ( 42 A- 42 E) of the image ( 14 ) onto the first photosensitive element ( 32 A- 32 E) light onto less than the entire first photosensitive element ( 32 A- 32 E) focused, and
the step of focusing light from a second portion ( 42 A- 42 E) of the image ( 14 ) onto the second photosensitive element ( 32 A- 32 E) light onto less than the entire second photosensitive element ( 32 A- 32 E) focused. 19. The method according to claim 17 or 18, wherein
the step of focusing light through the first and second lens elements ( 30 A- 30 E) inverts the first and second portions of the image ( 14 ), and
the step of combining the first and second signals comprises inverting the first and second signals to cancel the inversion of the image ( 14 ) and to resolve the overlapping areas. 20. The method according to any one of claims 17 to 19, the Step of illuminating the image.