JP4018092B2 - A device that automatically determines the cause of a given surface condition of an inspection object - Google Patents

Description

  The present invention relates to a technique for automatically determining a cause of a predetermined surface state according to a result of determining a predetermined surface state of an inspection object by image analysis.

  Inspection of scratches and other surface conditions on the surface of gears and other processed products is often performed by visual inspection. However, while the operation is simple, it requires concentration and is difficult. Therefore, there is a need for an apparatus that can inspect gears and other processed products without depending on visual observation.

Japanese Patent Application Laid-Open No. 2004-228561 describes a method of determining the presence or absence of a scratch such as a dent on the gear tooth surface by photographing the gear tooth surface and evaluating the luminance of the obtained image. Photographing a gear tooth surface to be inspected and a gear tooth surface different from the tooth surface with a predetermined interval for one gear, and creating a plurality of images from substantially the same direction regarding the different tooth surfaces. A difference image representing a difference in luminance is generated. The difference image is added to create an added image, and the presence or absence of a dent is determined by comparing the luminance of the added image with a reference value.
Patent Application Publication No. Sho 63-201556

  Gear scratches, dents, scratches such as peeling, and other predetermined surface conditions may be applied in various processes such as a manufacturing process and a storage process. If the cause of the generation of the predetermined surface state can be automatically determined from the detected scratches and other predetermined surface states, the management of gear production and storage can be improved efficiently.

  The present invention provides an apparatus capable of automatically determining the cause of a predetermined surface condition in response to detection of a scratch or other predetermined surface condition.

  In one form (Claim 1), the present invention is an apparatus that automatically determines the cause of the predetermined surface state in response to the determination of the predetermined surface state of the inspection object. The apparatus is obtained from a storage means for storing a cause pursuit map in which neurons are clustered for each cause of a predetermined surface state of an object in a self-organizing map, an imaging means for imaging an object, and an imaging means. Means for detecting a required determination area having a luminance difference exceeding a predetermined threshold with respect to another area in the image of the object, and specifying the inspection area so as to include the required determination area; A determination means for determining whether or not a surface state is included; a means for specifying position information of the inspection region in the object with respect to the inspection area determined to include a predetermined surface state; and Means for extracting a position vector representing the position of the surface state on the object, means for inputting the extracted position vector into the cause pursuit map, and a position vector input into the cause pursuit map Includes neurons distance between the combining vectors is minimized, means for identifying in the cause pursued map in accordance with the cluster Field of the identified neurons, means for determining the cause of the predetermined surface condition, the.

  According to the present invention, the cause of the predetermined surface state can be automatically obtained from the positional information of the inspection area determined to include the predetermined surface state. Since the cause of the predetermined surface state is automatically obtained, it is possible to efficiently determine which process related to the object should be improved.

  In one form of this invention (Claim 2), a target object is a gear and said position vector is represented by the position of the gear width direction of this gear.

  In one form of this invention (Claim 3), the position vector for learning in which the position on the target object in the predetermined surface state is specified for each target of the predetermined surface state is self-organized. A cause pursuit map is created by learning by inputting into the map.

  Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a configuration of an imaging apparatus when an object to be inspected is a gear 11. The gear 11 is attached to the rotary table 15 and rotated stepwise at a gear tooth pitch by a step motor (not shown) provided on the base 17. The gear 11 is irradiated by the lighting device 19. A camera 21 equipped with a CMOS image sensor photographs the tooth surface of the gear 11. Photographing is performed in synchronization with the stepwise rotation of the gear 11, and a still image is photographed for each tooth surface each time one tooth rotates. In one embodiment, the lighting device 19 uses blue light emitting diodes, but a lamp or other lighting device can also be used. Further, by using a wide dynamic range CMOS camera having a wide measurable luminance range as the camera 21, a good image suitable for scratch determination can be obtained. Depending on the object, a camera using a CCD image sensor can also be used.

  The positioning sensor 18 is generally called a proximity sensor, and generates a high-frequency magnetic field by a detection coil. When a metal detection object approaches this magnetic field, an induced current (eddy current) flows through the detection object due to electromagnetic induction. This current changes the impedance of the detection coil and stops the high-frequency oscillation, thereby detecting the approach of the detection object. In this embodiment, the proximity of the gear crest is detected by this proximity sensor. A signal from the positioning sensor 18 is sent to the computer 23 (FIG. 2), and the timing unit 24 sends a signal to the camera 21 in response to the gear tooth detection. The camera 21 shoots the gear in response to this signal.

  FIG. 2 shows functional blocks of the computer 23 that processes an image photographed by the camera 21. The computer 23 may be a personal computer, a workstation or other computer. The computer 23 includes a processor (CPU) that executes arithmetic operations, a storage device that stores various data and programs, an input device such as a mouse and a keyboard that allows a user to input data for processing, and the processor. A display device (display) for displaying the processing result and an output device such as a printer for outputting the processing result can be provided. The storage device can include a memory such as a ROM and a RAM and an auxiliary storage device such as a disk device. The processing by each functional block in the figure is executed by the processor.

  The image acquisition unit 25 receives an image captured for each tooth surface pitch from the camera 21 and passes it to the difference image generation unit 27. The difference image generation unit 27 calculates a difference between the image acquired this time and the image acquired last time, that is, the image of the gear tooth surface one pitch before. Since the tooth surfaces of the gear without any scratches or other surface conditions are substantially the same, taking a difference results in an image with a low overall brightness. If the gear tooth surface has scratches such as dents and peeling, the brightness of the scratched part is different from that of the other parts, so that it appears as a part of high brightness in the difference image.

  When the difference between the two images is taken, there is a pixel having a negative luminance value, so that the intermediate luminance 128 is added to the luminance value of the difference to make the luminance values of all the pixels positive. A pixel whose luminance value exceeds 255 by adding 128 has a luminance value of 255, and a pixel that becomes negative even if 128 is added has a luminance value of 0. In this way, the luminance value range of the difference image is adjusted from 0 to 255.

  Referring to FIG. 3A, the inspection area setting unit 28 detects an area having a luminance greater than the threshold value from the difference image as a potential flaw area 31, and sets an area 33 surrounding the area as an inspection area. To do. The shape of the inspection region 33 is a quadrangle in this embodiment, but may be other polygons, or may be a circle or an ellipse. The size is set to a size sufficient to surround the wound area 31 according to the potential wound area 31.

  Referring to FIG. 4, there is shown an example of three consecutive images 61 when the gear is rotated and the tooth surface of the gear is photographed for each tooth. Reference numeral 62 indicates a gear photographed in the image 61. The tooth width of the gear is represented by the length W. When the shooting cycle is represented by n, (A) in FIG. 4 is an image shot in the nth cycle, (B) is an image shot in the (n + 1) cycle, and (C) is (n + 2). ) Image taken at the cycle. In these images, an inspection area 33 is set. Since these three images are taken every time the gear rotates one tooth, the inspection region 33 extends over n, (n + 1), (n + 2) cycles, and h1, h2 in the height direction of the image. , H3, but the position in the tooth width direction is constant as w ′.

  Scratches and other surface conditions such as dents and delamination on the surface of the gear have a characteristic of reflecting illumination light in various directions, unlike over-detection such as color unevenness. Since scratches and other surface conditions have such characteristics, if images are acquired sequentially while rotating the gear relative to the camera, the scratches are located at the same position in the gear width direction across multiple consecutive images. Detected. In other words, if the inspection area includes a flaw, the inspection area is detected at the same position in the gear width direction of the gear over a plurality of consecutive images as shown in FIG. In this way, depending on whether the inspection region is set over a plurality of continuous images, it is possible to detect flaws and other surface conditions satisfactorily by distinguishing from overdetection such as color unevenness and dirt.

  The inspection area setting unit 28 maps the position information of the inspection area (in this example, the position in the tooth width direction and the cycle in which the inspection area is set) and the image of the inspection area, for example, as a map as shown in FIG. Can be recorded in the storage device of the computer 23. Referring to FIG. 5, the horizontal axis represents the position in the gear width direction of the gear, and the vertical axis represents the cycle. In this example, the gear tooth width W is equally divided into 20, and the position of the gear in the tooth width direction is specified by w0 to w19. In this example, since photographing is performed for each tooth and the number of gear teeth is 39, the cycle is specified by 0 to 38. The gear teeth can be specified by the cycle value. A1 to A4 indicate that the inspection region 33 is specified from the 30th cycle to the 34th cycle at the position w4 in the tooth width direction. B1 to B3 indicate that the inspection region 33 is specified from the 30th cycle to the 32nd cycle at the position w19 in the tooth width direction. C1 indicates that the inspection region 33 is specified in the sixteenth cycle at the position w10 in the tooth width direction. D1 and D2 indicate that the inspection region 33 is specified in the 16th cycle and the 17th cycle at the position w19 in the tooth width direction.

  Returning to FIG. 2, the first determination unit 29 refers to the information about the position and cycle of the examination region in the tooth width direction stored in the storage device. The first determination unit 29 continuously determines whether or not the inspection region 33 has been set a predetermined number of times (for example, three times) or more at the same position in the tooth width direction. If the inspection area 33 is set at the same position in the tooth width direction as in A1 to A4 and B1 to B3 at least the predetermined number of times, as described with reference to FIG. It is determined that the surface condition is included. Like C1 and D1 to D2, inspection areas other than the inspection area determined to be a scratch or other surface condition are determined to be inspection required areas. The determination result is stored for the inspection area registered in the storage device. Since it is necessary to determine whether or not the inspection-necessary region includes a flaw or other surface state in more detail, it is subjected to processing by the feature extraction unit 41 and the second determination unit 42.

  The feature extraction unit 41 extracts the following six parameters for the inspection region 33 determined as the inspection required region by the first determination unit 29.

(1) Area S _{A of} potential scratch region 31
The inspection area is displayed in coordinates with i as the vertical axis and j as the horizontal axis, and the area S _A is obtained by integration calculation using the coordinates of the boundary line of the scratch area 31.

(2) Long axis inclination θ _A with respect to the tooth streak when the wound region 31 is approximated to an ellipse
A dent often occurring in a gear can be approximated to an ellipse. In FIG. 3B, an approximate equation of an ellipse that approximates the flaw region 31 is obtained using a graphic approximation program. The approximate expression can be obtained using a known program. Typically, an approximate expression is obtained using a multivariate nonlinear least squares fitting program. An angle θ _A formed by the long axis 37 of the ellipse thus obtained and a straight line 35 parallel to the gear tooth stripe 13 (FIG. 1) is obtained.

(3) The luminance value entropy entB of the inspection area 33
The entropy of the luminance value of the inspection area 33 including the potential flaw area 31 is obtained by the following equation. The luminance value entropy represents the information amount of the luminance value distribution, and a large information amount represents that the degree of randomness of the luminance value distribution is large.

  In this embodiment, the randomness of the luminance value distribution from luminance values 0 to 255 is used as the luminance value entropy. In Equation 1, p is the luminance value, and rel [p] is the frequency of the luminance value p.

  FIG. 6 shows an example of a luminance value histogram. In the case of a gear tooth surface having dents or other scratches, since the scratches appear as pixel areas having various brightness values in the difference image, the brightness value entropy is large. FIG. 6A is a luminance value histogram of an image in which pixels having various luminance values are present at random. Brightness values are widely distributed around the intermediate brightness 128, and the entropy is large. (B) is a luminance value histogram of an image with small variations in luminance values, and has a small entropy.

(4) Luminance value anisotropy ansB of inspection area 33
The anisotropy of the luminance value of the inspection area 33 is expressed by the following formula. Anisotropy indicates the degree of symmetry of the luminance value distribution.

  In this expression, k is a luminance value with the lowest appearance frequency. That is, ansB is a value obtained by dividing the luminance value entropy from 0 to the minimum luminance value in the inspection region 33 by 0 to the entropy of all luminance values. When the luminance value distribution is symmetrical on the left and right of the intermediate luminance value 125, ansB = −0.5. The closer ansB is to -0.5, the higher the symmetry and the smaller the anisotropy. The closer ansB is to -0 or -1 from -0.5, the lower the symmetry and the greater the anisotropy.

  Referring to FIG. 7, the luminance value histogram as shown in FIG. 7A has a high luminance symmetry and a small anisotropy because the luminance values are distributed without any bias around the intermediate value 128. In the luminance value histogram of FIG. 7B, the luminance value distribution is different between the left side and the right side of the intermediate value 128, the symmetry is low, and the anisotropy is large. Assuming that the luminance value with the lowest appearance frequency is 230 in (B), ansB is close to -1.

(5) Circularity C _A of potential scratch area 31
Circularity C _A wound area 31 is expressed by the following equation.

In this equation, S _A is the area of the wound area 31, r _max is the maximum distance from the center of gravity of the wound area 31 to the end of the region 33.

(6) The average luminance edgB of the edge image in the inspection area 33
If represents the area of the inspection area 33 in S _B, the average luminance edgB edge images of the inspection area 33 is obtained by the following expression.

      Here, i and j indicate the coordinates of the pixel, and P (i, j) indicates the luminance value of the pixel at the coordinates i and j. ΔP (i, j) is the square of the difference between the luminance values of two pixels adjacent to the upper and lower sides (i-axis direction) of P (i, j) and the luminance of the two pixels adjacent to the left and right (j-axis direction) This is the square root of the sum of the difference squared. ΔP (i, j) represents the magnitude of the difference between the luminance values of adjacent pixels. Taking the difference between the luminance values of adjacent pixels corresponds to generating an image in which an edge portion having a large change in luminance value in the image is emphasized, that is, an edge image. edgB is obtained by dividing the sum of the magnitudes of the luminance values of adjacent pixels by the area of the region 33, and indicates the average luminance of the edge image.

  For each of the six parameters obtained in this way, normalization is performed according to the following equation using the maximum and minimum values for many samples in the inspection region.

Normalization = (obtained value-minimum value) / (maximum value-minimum value)
By this normalization, each parameter takes a value in the range of 0 to 1. Thus, the inspection region 33 including the potential flaw region 31 can be represented by a numerical vector of six feature parameters.

FIG. 8 shows the feature vector thus obtained. FIG. 8A shows a feature vector corresponding to a dent. FIG. 8B shows a feature vector of color unevenness (overdetection) of the gear tooth surface. Potential wound region area parameter S _A, the long axis of inclination theta _A when the elliptic approximation, and a large difference is not between the two is about anisotropy ansB, luminance values entropy entB, circularity C _A, and the edge image It can be seen that there is a large difference in the average luminance edgB.

  Returning to FIG. 2, the feature extraction unit 41 extracts feature vectors from the examination region 33 as described above. In one embodiment, the feature extraction unit 41 extracts a feature vector composed of the six parameters described above. In an alternative embodiment, the feature extraction unit 41 extracts a feature vector composed of three parameters: the luminance value entropy of the inspection region, the circularity of the potential flaw region, and the average luminance of the edge image of the inspection region.

  The second determination unit 42 calculates the distance between the feature vector of the flaw model generated based on the test with a large number of samples in advance and the feature vector extracted by the feature extraction unit 41, and the distance is a determination threshold value. When there is a smaller scratch model, it is determined that there is a scratch. When there is no scratch model whose distance is smaller than the determination threshold, it is determined that the inspection area includes not a scratch but overdetection. The determination result is stored for the inspection area registered in the storage device.

In an alternative embodiment, the second determination unit 42 generates a self-organizing map using neurons and determines a wound using this map. This alternative embodiment will be described. In this embodiment, 40 × 40 neurons are used to generate a self-organizing map. FIG. 9 shows three neurons N _j−1 , N _j and N _{j + 1} . ω _j = (ω _j0 , ω _j1 ,..., ω _jm ) is a coupling coefficient vector of the neuron N _j , and χ = (χ ₀ , χ ₁ ,..., χ _m ) is an input vector is there. The self-organizing map is also called a self-organizing feature map and is an algorithm without learning. This map extracts hidden features in the input data and automatically learns to generate a map that selectively reacts to the input data. The self-organizing map was proposed by Kohonen and explained in, for example, “Neuro Fuzzy Genetic Algorithm” published in 1994 by Sangyo Tosho Co., Ltd.

  The input vector in this embodiment is a feature vector extracted by the feature extraction unit 41 (FIG. 2) for a plurality of sample inspection areas 33. A self-organizing map is created in the next step.

1) Step 1: Network initialization:
The neuron connection vector ω _j = (ω _j0 , ω _j1 ,..., Ω _jm ) is determined for all neurons using random numbers.

2) Step 2: Input vector input:
An input vector χ = (χ ₀ , χ ₁ ,..., Χ _m ) is given to each neuron. The relationship between each neuron and the input vector is as shown in FIG.

3) Step 3: Calculate distance between neuron coupling coefficient vector and input vector:
The distance between the connection vector of each neuron and the input vector is calculated by the following equation.

4) Step 4: Determine winner neuron:
The neuron with the smallest distance d _j is selected. This neuron is called the winner neuron and is represented by j *.

5) Step 5: Learning coupling coefficient vectors:
The coupling coefficients (weights) of the winner neuron and nearby neurons are updated according to the following formula.

η is a positive constant and is 0.05 in this embodiment. h (j, j *) is called a neighborhood function and is represented by the following equation.

  σ (t) decreases as the learning progresses. Therefore, the range of the neighborhood function is wide at the initial stage of learning, as indicated by a circle in FIG. 10, and becomes narrower as the learning progresses. In other words, the coarse adjustment changes to the fine adjustment as the learning progresses. Thus, the neighborhood function effectively generates a map. FIG. 10 shows the winner neuron as a small circle 91, and the neighborhood function surrounding it is represented as a large circle 93.

6) Step 6: t update, return to input vector input:
The learning count t is updated to t + 1, the input vector input in step 2 is returned, and steps 2 to 6 are repeated to repeatedly update the coupling coefficient vector.

  In the self-organizing map, the winner neuron and nearby neurons all approach the current input vector. At the beginning of learning, many neurons are considered to be close by the neighborhood function, and a rough map is formed. As the learning progresses, the number of neurons considered to be close to the winner neuron by the neighborhood function decreases. For this reason, local fine adjustment progresses and spatial resolution increases.

  Features obtained by learning by selecting all the feature vectors of the examination area 33 including scratches and other surface states or over-detection obtained for a plurality of samples in order or randomly and inputting them into a self-organizing map A self-organizing map reflecting the similarity of vectors can be generated. That is, a cluster having similar characteristics gathers on the self-organizing map. FIG. 11 shows the self-organizing map obtained in this way. In the embodiment, 40 × 40 neurons are used, but FIG. 11 shows a 20 × 20 map for simplicity. In FIG. 11, each cell corresponds to one neuron. Blocks (1), (2),... (28) separated by solid lines represent clusters determined as follows.

  An image of the examination region having a feature vector having the smallest distance from the coupling coefficient vector is arranged at each neuron position of the self-organizing map. This examination area is one of a plurality of samples used for generating the self-organizing map. Adjacent neurons with the same image are selected and grouped. A group of neurons formed in this way is called a cluster. The grouping here is the same as the grouping function used in the existing drawing program, and the grouped neurons can be selected as a whole group and the properties can be set.

  Next, the images arranged in the respective clusters are visually confirmed, and it is determined whether the inspection region 33 includes a scratch or other surface condition or an overdetection such as surface unevenness. This determination result is set as a cluster property. Specifically, with “Overdetection” entered in the property fields of all clusters as the initial state, a self-organizing map is displayed on the computer display, and the clusters determined to be scratched are displayed with the mouse. Right-click and change and enter “Scratch”. Since neurons can be classified in cluster units in this way, working time can be shortened.

  FIG. 12 shows a map (determination data map) in which clusters are classified in this way. In the figure, hatched clusters (1), (2), (6), (10), (11), (12), (13), (15), (16), (22), (27 ), (28) are clusters classified as “scratches”, and clusters without hatching are “overdetected” clusters. This determination data map is registered in the storage device of the computer 23 (FIG. 2), and this is used by the second determination unit 42 for determination of scratches.

  A specific method for determining scratches performed by the second determination unit 42 will be described. The second determination unit 42 obtains the distance between the feature vector received from the feature extraction unit 41 and the coupling coefficient vector of the determination data map. This distance is calculated according to [Equation 5] shown above. Neurons are selected in ascending order of the distance, a predetermined number, for example, 10 or less neurons are selected, and a circle or ellipse surrounding them is set as a neighborhood region. As is clear from the generation process of the decision data map, the neurons that approximate the coupling coefficient vector are grouped as a cluster, so if you select a neuron in ascending order of distance as described above, a cluster of adjacent neurons is formed. Selected. A circle or ellipse surrounding this cluster of neurons is defined as a neighborhood region. For example, referring to FIG. 13, a circle 57 defines the neighborhood region.

  A circle 57 obtains the center position of the cluster of these neurons from the array on the determination data map of the selected cluster of neurons, and a circle having a radius from this center to the farthest neuron of the cluster of neurons. Can be obtained as

  Based on the ratio (K) / (S) of the number of neurons (K) belonging to the wound cluster included in the neighboring area defined by the circle or ellipse thus selected and the total number of neurons (S) in the neighboring area Check for scratches. When (K) / (S) ≧ D, it is judged as a scratch, and when (K) / (S) <D, it is judged as overdetection. D is a preset threshold value, for example, 0.5. In this way, the scratch and other surface conditions and overdetection are discriminated. In the example of FIG. 13, since there are seven neurons belonging to the wound cluster (12) in the circle 57 and the total number of neurons in the circle 57 is 12, (K) / (S) = 8/12 = Therefore, it is determined that a flaw is included in the current examination region 33. The determination result is stored for the inspection area registered in the storage device.

  Returning to FIG. 2, the wound position vector extraction unit 45 refers to the position in the tooth width direction of the inspection region determined to be a scratch or other surface state from the storage device, and calculates a wound position vector for each gear. The flaw position vector indicates at which position in the tooth width direction the inspection region determined to include a flaw or other surface condition is represented by the following equation.

Scratch position vector = (w0, w1, w2,…, w19)
As described above, the vector components w0 to w19 represent positions in the gear width direction of the gear. A value 1 is set to the vector component corresponding to the position where the scratch or other surface condition is determined, and zero is set to the remaining vector components. For example, in the example of the gear shown in FIG. 5, since the scratch is at the position of w4 and w19, the scratch position vector is (0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 , 0, 0, 0, 0, 0, 0, 0, 0, 1).

  The cause pursuit unit 46 inputs the flaw position vector into the cause pursuit map. The cause pursuit map is a self-organizing map in which neurons are clustered for each cause of occurrence of scratches and other surface conditions, and is stored in the storage device of the computer 23. An example of the cause pursuit map is shown in FIG. Each cell represents a neuron. In this example, the production process cluster that causes the scratch in the gear production process, the lamination process cluster that causes the scratch in the lamination process that stacks the produced gear, the transportation process cluster that causes the scratch in the process that transports the gear, Neurons are clustered (clustered) in a storage process cluster in which the cause of scratches is in the storage process of storing gears in a warehouse or the like.

  The cause pursuit unit 46 determines the cause by the same method as the above-described determination described in the alternative embodiment of the second determination unit 42. Specifically, the cause pursuit unit 46 obtains the distance between the flaw vector extracted by the flaw position vector extraction unit 45 and the connection vector of each neuron of the cause pursuit map. This distance is calculated according to the same method as described with reference to [Equation 5]. The neuron with the minimum distance is extracted. A cluster to which the extracted neuron belongs is determined, and this cluster is determined as a process in which a scratch or other surface condition has occurred. For example, if the neuron having the smallest distance between the flaw position vector and the coupling vector for a certain gear is the neuron indicated by reference numeral 59 (FIG. 14), it is estimated that the gear flaw has occurred in the production process. be able to.

  Next, a method for creating a cause pursuit map is described. The cause pursuit map is created using a scratch position vector for learning as an input. An example of the flaw position vector for learning is shown in FIG. These vectors are prepared in advance by examining the relationship between the positions of scratches and other surface states and the causes of the occurrence of a plurality of gears. For example, the laminating step is a step of stacking gears, and it has been found that one edge (end) portion of the gear, that is, the portion of w16 to w19 in the tooth width direction is easily damaged. Therefore, the learning flaw position vector for the stacking process is set to 1 in the vector components w16 to w19.

  The method of creating a self-organizing map with the learning flaw position vector as an input is the same as the method of creating a self-organizing map for the determination data map. That is, after the network initialization in step 1 is executed, step 2 to step 6 are repeated. As the learning flaw position vector is repeatedly input to the self-organizing map for learning, the range of neurons that respond to each learning flaw position vector is defined by clusters as shown in FIG. In this way, a cause pursuit map in which neurons are clustered for each cause of generation of scratches and other surface states is generated. The generated cause pursuit map is registered in the storage device.

  FIG. 16 is a flowchart showing a process for pursuing the cause of the surface condition in response to the determination of a gear scratch or other surface condition in one embodiment of the present invention. The gear 11 is attached to the rotary table of FIG. In step S101, it is determined whether or not all gear tooth surfaces of the gear 11 have been inspected. If incomplete, the process proceeds to step S103. To determine whether all gear tooth surfaces have been inspected, the number of teeth of the gear 11 is set in the counter, and the counter is decremented each time processing for one tooth is completed, and the value of this counter is checked. By doing.

  In step S103, the gear 11 is rotated by one tooth, stopped at the position where the positioning sensor 18 is turned on, and positioned. The positioning sensor 18 is set to output a signal when the rotational position of the gear 11 reaches a predetermined position.

  In step S105, the gear 11 is photographed, and the image is sent to the image processing personal computer 23. When the number of photographing is 1 (S107), that is, when the first tooth of the gear 11 has just been photographed, image processing using the difference image cannot be performed, so this image is saved and the next tooth is stored. The process returns to step S101 for shooting.

  In step S109, as described above, a difference image is generated from two consecutive tooth images, and an examination region 33 surrounding the potential wound region 31 (FIG. 3) is set. Next, in step S111, for the set inspection region 33, the position of the gear in the tooth width direction, the value of the counter (which identifies the cycle), and the image of the inspection region are stored. Then, it returns to step S101.

  Thus, when the determination in step S101 is No, the position in the tooth width direction of the inspection region, the cycle in which the inspection region is set, and the image of the inspection region are stored for all teeth of the gear.

  In step S113, it is checked whether or not there is an inspection region set at the same position in the tooth width direction over a predetermined number of consecutive cycles with reference to the information on the inspection region stored in the storage device in step S111. . If it exists, it is determined that the inside of the inspection region includes a scratch or other surface condition (first determination). In step S111, information related to the inspection area stored in the storage device is updated so that the inspection area determined to include a scratch or other surface condition is identified.

  In step S115, information on the inspection area stored in the storage device is referred to, and it is checked whether or not there is an inspection area other than the inspection area determined to include a scratch or other surface condition, that is, an inspection required area. If the inspection-necessary area exists, a feature vector is extracted from the inspection-necessary area in step S117, and in step S119, whether the inspection-necessary area includes a flaw or other surface state using the feature extraction vector. Is determined (second determination). The second determination may be any of the two embodiments described above. The information regarding the inspection area stored in the storage device in step S111 is updated so that the inspection area determined to include scratches and other surface conditions is identified again. In step S121, a routine (FIG. 17) for calculating the cause of scratches and other surface conditions is executed.

  FIG. 17 is a flowchart of a process for calculating the cause of a scratch or other surface condition, which is executed in step S121 of FIG. In step S131, a cause pursuit map as shown in FIG. 14 is read from the storage device. In step S132, the information on the inspection area of the storage device is referred to, and a flaw position vector is calculated from the position in the tooth width direction of the inspection area determined to be a flaw or other surface state in steps S113 and S119 in FIG. In step S133, the distance between the wound position vector and the connection vector of each neuron in the cause pursuit map is obtained, and the neuron with the minimum distance is specified in the cause pursuit map. In step S134, a cluster to which the extracted neuron belongs is determined, and this cluster is determined to be a process in which a scratch or other surface condition has occurred.

  In the above embodiment, in the first stage, scratches and other surface state characteristics are used to determine the scratches by a simpler method. For the inspection area that cannot be determined by the first determination, a feature vector is used. By carrying out the determination with the finer technique used, the determination can be performed efficiently while improving the accuracy of the determination. However, alternatively, the flaw position vector may be calculated and input to the cause pursuit map only for the flaw and other surface states determined by the first determination unit. Alternatively, scratches and other surface states may be determined by the second determination for all inspection regions, and a scratch position vector may be extracted for the determined scratches and other surface states.

  Although the present invention has been described with respect to specific embodiments, the present invention is not limited to such embodiments.

The block diagram which shows the whole structure of the imaging device of a gear tooth surface according to one Example of this invention. 1 is a functional block diagram showing an overall configuration of an image processing computer according to one embodiment of the present invention. The figure which shows the relationship between a potential flaw area | region and a test | inspection area | region according to one Example of this invention. The figure which shows the method of determining a damage | wound according to 1st determination according to one Example of this invention. The figure which shows the map produced | generated from the set test | inspection area | region according to one Example of this invention. The figure which shows the luminance value histogram from which entropy differs according to one Example of this invention. The figure which shows the luminance value histogram from which anisotropy differs according to one Example of this invention. The figure which shows the feature vector according to one Example of this invention. The figure which shows the partial structure of the neural network according to one Example of this invention. The figure which shows the production | generation process of the self-organization map according to one Example of this invention. The figure which shows the self-organization map divided into the clusters according to one Example of this invention. The figure which shows the determination data map produced | generated from the self-organization map according to one Example of this invention. The figure for demonstrating the damage | wound determination using the determination data map according to one Example of this invention. The figure which shows the cause pursuit map according to one Example of this invention. The figure which shows the flaw position vector for learning according to one Example of this invention. The flowchart which shows the main routine of the process which determines the damage | wound of a gear tooth surface and other surface states and calculates | requires a cause according to one Example of this invention. The flowchart which shows the subroutine of the process which calculates | requires the cause according to one Example of this invention.

Explanation of symbols

11 Gear 19 Lighting device (LED)
21 Camera 23 Computer

Claims (6)

In response to the determination of the predetermined surface state in the inspection object, an apparatus that automatically determines the cause of the predetermined surface state,
Storage means for storing a cause seeking map in which neurons are clustered for each cause of a predetermined surface state of the object in a self-organizing map;
Imaging means for imaging the object;
Means for detecting a required determination area having a luminance difference exceeding a predetermined threshold with respect to another area in the image of the object acquired from the imaging means, and specifying an inspection area to include the required determination area When,
Determining means for determining whether the inspection region includes the predetermined surface state;
Means for specifying position information of the inspection region in the object for the inspection region determined to include the predetermined surface state;
Means for extracting from the position information a position vector representing the position of the predetermined surface state on the object;
Means for inputting the extracted position vector into the cause pursuit map;
Means for identifying in the cause pursuit map a neuron in which the distance between the position vector input to the cause pursuit map and the coupling vector is minimum;
Means for determining a cause of the determined predetermined surface state according to a cluster to which the identified neuron belongs;
An apparatus comprising: The object is a gear, and the position vector is represented by a position of the gear in the tooth width direction.
The apparatus of claim 1. Further, the learning position vector in which the position of the predetermined surface state on the target object is specified for each cause of the predetermined surface state is input to the self-organizing map to be learned. A means for creating the cause pursuit map.
The apparatus of claim 1. A program for causing a computer to automatically determine the cause of the predetermined surface state according to the determination of the predetermined surface state in the inspection object,
The computer includes a storage device that stores a cause search map in which neurons are clustered for each cause of a predetermined surface state of the object in a self-organizing map;
In the image of the object acquired from the imaging device that images the object, a required determination area having a luminance difference exceeding a predetermined threshold with respect to another area is detected, and an inspection is performed to include the required determination area The ability to identify areas,
A determination function for determining whether the inspection region includes the predetermined surface state;
For the inspection area determined to include the predetermined surface state, a function for specifying position information on the object of the inspection area;
A function of extracting a position vector representing the position of the predetermined surface state on the object from the position information;
A function of inputting the extracted position vector into the cause pursuit map;
A function for specifying in the cause pursuit map a neuron that minimizes the distance between the position vector input to the cause pursuit map and the coupling vector;
A function for determining the cause of the determined predetermined surface state according to the cluster to which the identified neuron belongs;
A computer program that realizes The object is a gear, and the position vector is represented by a position of the gear in the tooth width direction.
The computer program according to claim 4. Further, the learning position vector in which the position of the predetermined surface state on the target object is specified for each cause of the predetermined surface state is input to the self-organizing map to be learned. By providing a function to create the cause pursuit map,
The computer program according to claim 4.

Images