JP2023153431A - Image processing device, inspection system, and inspection method - Google Patents

Abstract

To provide an image processing device, an inspection system, and an inspection method capable of reducing a user's workload.SOLUTION: An image processing device 28 processes image data for an inspection object that has been photographed under illumination. An illumination condition searcher 26 variably sets an illumination condition ILC[i] for the illumination. An image interface 25 acquires image data IMD[i] for each illumination condition ILC[i]. An image analyzer 27 carries out image analysis on the image data IMD[i] for each illumination condition ILC[i]. A comparator 28 compares an image analysis result AR[i] from the image analyzer 27 for each illumination condition ILC[i] to select any one illumination condition ILCx of a plurality of illumination conditions ILC[i].SELECTED DRAWING: Figure 2

Description

The present invention relates to an image processing device, an inspection system, and an inspection method, and for example, to a technique for performing an inspection based on an image of an object to be inspected.

For example, Patent Document 1 discloses a lighting device that includes a projector, a translucent liquid crystal display, an imaging device, and an image processing device, and that enables high-definition lighting according to the shape and color of an object to be illuminated. . Specifically, the irradiated object is irradiated with light from a projector via a translucent liquid crystal display, and the imaging device section images the irradiated object. The image processing device processes the video signal from the imaging device section and distinguishes between an area of the irradiated object and an area other than the irradiated object. The image processing device then controls the translucent liquid crystal display to mask irradiation to areas other than the irradiated object.

Furthermore, Patent Document 2 discloses a lighting control device that includes a CCD sensor that can form an electrical image of a room, and a control means that controls indoor lighting according to the measurement results of the CCD sensor.

Japanese Patent Application Publication No. 08-271985 Special Publication No. 2004-501496

For example, an inspection system is known that includes an imaging device (typically a camera) that captures an image of an object to be inspected, and an illumination device that irradiates the object with light to assist the imaging by the camera. The lighting device includes, for example, a halogen lamp light source, an LED light source, etc., and is capable of adjusting lighting conditions such as brightness and color (R, G, B, IR, etc.) so that a clear image of the object to be inspected can be captured. It has become. However, adjustment of the lighting conditions is typically done manually by the user of the device. For this reason, when deriving the optimal value of the lighting conditions, there is a risk that the user's workload and working time will increase.

The present invention has been made in view of the above, and one of its purposes is to provide an image processing device, an inspection system, and an inspection method that can reduce the workload of the user. .

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

In one embodiment of the present invention, an image processing device is provided that processes image data of an inspection target imaged while being illuminated, and the image processing device is an illumination condition searcher that variably sets the illumination conditions of the illumination. By comparing the image analysis results for each lighting condition, an image interface that acquires image data for each lighting condition, an image analyzer that performs image analysis on the image data for each lighting condition, and an image analyzer for each lighting condition. A comparator that selects any one illumination condition from among a plurality of illumination conditions may be configured.

Of the inventions disclosed in this application, the effects obtained by typical inventions will be briefly described, making it possible to reduce the user's workload.

1 is a schematic diagram showing a configuration example of main parts in an inspection system according to Embodiment 1 of the present invention. FIG. 2 is a block diagram showing a configuration example of main parts of the image processing device in FIG. 1. FIG. 3 is a flow diagram showing an example of processing contents of the image processing apparatus of FIG. 2. FIG. 3 is a schematic diagram illustrating an example of the operation of an image analyzer and a comparator in FIG. 2. FIG. 5 is a flow diagram showing an example of processing contents of an image analyzer and a comparator in FIGS. 2 and 4. FIG. FIG. 5A is a supplementary figure to FIG. 5A. 3 is a schematic diagram illustrating an example of the operation of an image analyzer and a comparator in FIG. 2 in the image processing apparatus according to Embodiment 2 of the present invention. FIG. FIG. 6 is a diagram illustrating an example of an edge extraction method using a Sobel filter by an image analyzer. 3 is a schematic diagram showing a configuration example of an image analyzer in FIG. 2 in an image processing apparatus according to Embodiment 3 of the present invention. FIG.

Hereinafter, embodiments of the present invention will be described in detail based on the drawings. In addition, in all the figures for explaining the embodiment, the same members are given the same reference numerals in principle, and repeated explanations thereof will be omitted.

(Embodiment 1)
《Inspection system overview》
FIG. 1 is a schematic diagram showing a configuration example of main parts in an inspection system according to Embodiment 1 of the present invention. The inspection system 10 shown in FIG. 1 includes an imaging device 15, a lighting device 16, a lighting control device 17, and an image processing device 18. The illumination device 16 includes, for example, a halogen lamp light source, an LED (Light Emitting Diode) light source, and the like, and illuminates the inspection object 20a so that a clear image of the inspection object 20a can be captured.

The imaging device 15 includes, for example, a CMOS (Complementary Metal Oxide Semiconductor) sensor or a CCD (Charge Coupled Device) sensor, and creates image data IMD by capturing an image of the inspection target 20a in a state where it is illuminated. The image processing device 18 acquires image data IMD from the imaging device 15 via the cable 19a, and processes the image data IMD. As an example, the image processing device 18 inspects the mounting state (for example, the mounting position of the IC, the pitch and number of IC leads, etc.) of the electronic component (IC) 20a mounted on the wiring board 20b by image processing. In this case, in the image data IMD, the wiring board 20b becomes the background, and the electronic component 20a becomes the object to be inspected.

Furthermore, although details will be described later, the image processing device 18 searches for the optimal lighting condition ILC by variably setting the lighting condition ILC for the lighting device 16 and processing the corresponding image data IMD. The lighting control device 17 receives the lighting condition ILC from the image processing device 18 via the cable 19b, and transmits the lighting control signal ICS corresponding to the lighting condition ILC. The lighting device 16 receives the lighting control signal ICS from the lighting control device 17 via the cable 19c, generates light based on the lighting control signal ICS, and irradiates the object to be inspected 20a with the light. Note that the lighting control device 17 may be integrated with the image processing device 18 or the lighting device 16.

Here, in the general inspection system 10, the cable 19b is not provided, and the lighting conditions ILC are manually set in the lighting control device 17 by the user. That is, for example, the user visually evaluates the sharpness of the corresponding image data IMD for each type of inspection object 20a while appropriately changing the illumination condition ILC prior to inspecting the inspection object 20a. Then, the user derives the illumination condition ILC that provides the highest clarity. However, such a method may increase the user's workload and work time.

For example, when the illumination device 16 uses a monochromatic light source and the light amount can be set in 256 gradations, there are 256 options as the illumination condition ILC. Further, when the illumination device 16 uses three color light sources (R, G, B) and the light amount can be set in 16 gradations for each, there are 4096 (=16 ³ ) choices as the illumination condition ILC. The user needs to visually compare the image data IMD obtained for each illumination condition ILC option. Furthermore, in addition to such workload issues, user proficiency is also required in order to obtain the optimal value of the illumination condition ILC. Therefore, it is advantageous to use the image processing device 18 of the embodiment.

《Details of image processing device》
FIG. 2 is a block diagram showing an example of the configuration of main parts of the image processing apparatus in FIG. 1. As shown in FIG. The image processing device 18 shown in FIG. 2 includes an image interface 25, an illumination condition searcher 26, an image analyzer 27, a comparator 28, and a memory 29. The illumination condition searcher 26 variably sets the illumination condition ILC[i] of the illumination device 16. The image interface 25 acquires image data IMD[i] of the inspection object 20a imaged and created by the imaging device 15 for each illumination condition ILC[i], and converts the image data IMD[i] to the illumination condition ILC[i]. i] and stored in the memory 29.

The image analyzer 27 performs image analysis on the image data IMD[i] for each illumination condition ILC[i] stored in the memory 29, and applies the image analysis result AR[i] to the illumination condition ILC[i]. The data are stored in the memory 29 in association with each other. The comparator 28 compares the image analysis results AR[i] for each illumination condition ILC[i] stored in the memory 29 to select one illumination condition ILCx from among the plurality of illumination conditions ILC[i]. Select. Then, the comparator 28 sets the selected illumination condition ILCx as the final illumination condition ILCx when inspecting the inspection object 20a.

Note that the image processing device 18 in FIG. 2 includes, for example, one or more processors (CPU (Central Processing Unit) or GPU (Graphics Processing Unit)) and memory 29 (RAM (Random access memory) and nonvolatile access memory) on a motherboard. It is configured in a form that implements a physical memory). In this case, the image interface 25, the illumination condition searcher 26, the image analyzer 27, and the comparator 28 are implemented by program processing using such a processor.

However, the image processing device 18 is not limited to this implementation form, but may also be a form using an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), etc. instead of a processor, or a PC (Personal Computer). It may also be a form using etc. Although not shown in the drawings, the image processing device 18 is additionally equipped with an image analyzer for, for example, performing image analysis associated with the inspection of the inspection object 20a and making a quality determination.

FIG. 3 is a flow diagram illustrating an example of processing contents of the image processing apparatus of FIG. 2. The flow in FIG. 3 is executed, for example, for each type of object 20a to be inspected, prior to the inspection of the object 20a to be inspected. In FIG. 3, first, the illumination condition searcher 26 sets the illumination condition ILC[i] to an initial value (step S101). Subsequently, the image interface 25 acquires image data IMD[i] of the inspection object 20a imaged and created by the imaging device 15 under the illumination condition ILC[i] (step S102). Next, the image analyzer 27 performs image analysis on the acquired image data IMD[i] (step S103), associates the image analysis result AR[i] with the illumination condition ILC[i], and stores it in the memory. 29 (step S104).

Subsequently, the illumination condition searcher 26 determines whether there is another illumination condition ILC[i] (step S105). If there is another illumination condition ILC[i], the illumination condition searcher 26 changes the illumination condition ILC[i] (step S106). After that, similarly to steps S102 to S104, the image interface 25 acquires the image data IMD[i] (step S107), and the image analyzer 27 performs image analysis (step S108). The result AR[i] is stored in the memory 29 (step S109). Then, the image processing device 18 repeatedly executes the processes of steps S106 to S109 until there is no other illumination condition ILC[i] in step S105.

Thereby, image analysis results AR[i] are obtained as many as the number of options for illumination condition ILC[i]. For example, as mentioned above, when there are 256 options, 256 image analysis results AR[i] are obtained, and when there are 4096 options, 4096 image analysis results AR[i] are obtained. can get. Note that if the lighting device 16 is designed to adjust the amount of light in an analog manner, for example, the analog value may be quantized into a plurality of stages and then similar processing may be performed.

On the other hand, if it is determined in step S105 that there is no other illumination condition ILC[i], the comparator 28 compares the image analysis results AR[i] for each of the plurality of illumination conditions ILC[i] (step S110). . Then, the comparator 28 determines the illumination condition ILC[i] that provides the best image analysis result AR[i] as the final illumination condition ILCx (step S111). Thereafter, the inspection object 20a is inspected under this final illumination condition ILCx.

Note that the flow of the image processing device 18 is not limited to the flow shown in FIG. 3, and can be changed as appropriate. For example, in the loop processing of steps S105 to S109, the previous image analysis result (AR[i-1]) and the current image analysis result (AR[i]) are compared, and whichever is better is selected. The flow may be such that only the analysis results (and the corresponding illumination conditions) are left in the memory 29.

《Details of image analyzer》
FIG. 4 is a schematic diagram illustrating an example of the operation of the image analyzer and comparator in FIG. 2. In FIG. 4, the image analyzer 27 calculates contrast for image data 31a (IMD) and 31b (IMD) for each illumination condition ILC[1] and ILC[2]. Specifically, the image analyzer 27 calculates the contrast between the density of the inspection object 20a and the density of the background 20b for each image data 31a (IMD) and 31b (IMD). In this example, for convenience of explanation, it is assumed that the inspection object 20a has a circular shape. Then, the image analyzer 27 outputs the contrast calculated for each image data 31a (IMD) and 31b (IMD) as image analysis results AR[1] and AR[2].

The comparator 28 compares the image analysis results AR[1] and AR[2] (i.e. contrast) to determine which one of the plurality of illumination conditions ILC[1] and ILC[2]. Select. In the example of FIG. 4, the image data 31b under illumination condition ILC[2] has a higher contrast than the image data 31a under illumination condition ILC[1]. In this case, the comparator 28 determines that the illumination condition ILC[2] with large contrast is better than the illumination condition ILC[1], and selects the illumination condition ILC[2].

FIG. 5A is a flow diagram illustrating an example of processing contents of the image analyzer and comparator in FIGS. 2 and 4. Figure 5B is a supplementary figure to Figure 5A. One of the methods for calculating the contrast described in FIG. 4 is a method using a discriminant analysis method. The discriminant analysis method is also called Otsu's binarization, and as shown in FIG. 5B, is a method of determining a threshold value TH for binarizing (black-and-white) all pixels in an image.

FIG. 5B shows a histogram representing the relationship between pixel values (density) and the number of pixels for each pixel value. In order to binarize an image with such a histogram, a group of pixels with pixel values less than or equal to the threshold TH are classified into class [1] (i.e. one of 0/1), and a group of pixels larger than the threshold TH is classified into class [1] (i.e. one of 0/1). A pixel group having a pixel value is classified into class [2] (the other of 0/1). In the discriminant analysis method, the threshold value TH with which the degree of separation S is maximized is searched for by calculating the degree of separation S shown in equation (1) while changing the threshold value TH.
S=σ _b ² /σ _w ² …(1)

In equation (1), σ _b ² is the inter-class variance and is given by equation (2). σ _w ² is the intra-class variance and is given by equation (3). In addition, in equations (2) and (3), w ₁ , m ₁ , σ ₁ are the number of pixels, the average value of pixel values, and the variance of class [1], respectively, and w ₂ , m ₂ , σ ₂ are the number of pixels, the average value of pixel values, and the variance of class [2], respectively.
σ _b ² = w ₁ w ₂ (m ₁ - m ₂ ) ² / (w ₁ + w ₂ ) ² ...(2)
σ _w ² = (w ₁ σ ₁ ² + w ₂ σ ₂ ² )/(w ₁ +w ₂ ) …(3)

Here, as can be seen from the numerator of equation (2), the degree of separation S in equation (1) increases as the difference between the average value m ₁ of class [1] and the average value m ₂ of class [2] increases. Become. The difference between these average values represents the contrast. That is, when the discriminant analysis method is used, by calculating the maximum degree of separation S for each image, it becomes possible to indirectly calculate the contrast index value for each image.

Using such a discriminant analysis method, the image analyzer 27 and the comparator 28 perform processing as shown in FIG. 5A. First, the image analyzer 27 stores the degree of separation S (maximum degree of separation Smax) calculated using the discriminant analysis method in the memory 29 as the image analysis result AR[i] for each illumination condition ILC[i] ( Step S201). After that, the comparator 28 compares the degree of separation (Smax) stored in the memory 29 (step S202). Then, the comparator 28 selects the illumination condition where the degree of separation (Smax) is maximized (in other words, the contrast is maximized) as the final illumination condition ILCx (step S203).

《Main effects of Embodiment 1》
As described above, by using the method of Embodiment 1, it is typically possible to reduce the user's workload associated with determining lighting conditions. Moreover, it becomes possible to shorten the user's working time, and by extension, it becomes possible to shorten the inspection time. Furthermore, the user's level of proficiency is not required when determining the illumination conditions, and it becomes possible to automatically determine the illumination conditions based on an objective index (in this case, contrast). As a result, it becomes possible to suppress variations in inspection accuracy.

(Embodiment 2)
《Details of image analyzer》
FIG. 6 is a schematic diagram illustrating an example of the operation of the image analyzer and comparator in FIG. 2 in the image processing apparatus according to the second embodiment of the present invention. In FIG. 6, the image analyzer 27 uses an edge extraction filter for image data 31c (IMD) and 31d (IMD) for each illumination condition ILC[3] and ILC[4] to Edges of the object 20a) are extracted, and image data 32c and 32d after edge extraction are created. Furthermore, the image analyzer 27 calculates intensity values E[3] and E[4] of the extracted edges for the image data 32c and 32d. Then, the image analyzer 27 outputs the edge strength values E[3] and E[4] as image analysis results AR[3] and AR[4].

The comparator 28 compares the image analysis results AR[3] and AR[4] (that is, the edge intensity values E[3] and E[4]), thereby determining the illumination conditions ILC[3] and ILC. Select one of the lighting conditions from [4]. FIG. 6 shows the relationship between pixel positions (X coordinates) and values (cumulative pixel values) obtained by accumulating pixel values (densities) in the Y coordinate direction for each X coordinate in the image data 32c and 32d. The edge strength values E[3] and E[4] are determined, for example, by the peak value of the cumulative pixel values.

In this example, the edge intensity value E[4] in the image data 32d under the illumination condition ILC[4] is higher than the edge intensity value E[3] in the image data 32c under the illumination condition ILC[3]. In this case, the comparator 28 determines that the illumination condition ILC[4] with a high edge intensity value is better than the illumination condition ILC[3], and selects the illumination condition ILC[4]. Here, edge intensity values E[3] and E[4] were used as the image analysis results AR[3] and AR[4], but in addition, the sum of cumulative pixel values, etc. It's okay. Specifically, for example, a value obtained by performing a sum-of-products operation on the edge strength value and the total sum of cumulative pixel values with predetermined weighting may be used.

Here, a typical example of the edge extraction filter used in FIG. 6 is a Sobel filter. FIG. 7 is a diagram illustrating an example of an edge extraction method using a Sobel filter by the image analyzer in FIG. 6. In the Sobel filter, as shown in FIG. 7, a kernel Kx of 3 rows and 3 columns is used to extract edges in the row direction, and a kernel Ky of 3 rows and 3 columns is used to extract edges in the column direction. It will be done.

Further, the image data IMD includes pixel values (densities) I[11] to I[xy] in x row and y column. The image analyzer 27 creates image data in which edges in the row direction are extracted by convolving the image data IMD with the kernel Kx. Similarly, the image analyzer 27 creates image data in which edges in the Y direction are extracted by convolving the image data IMD with the kernel Ky.

Specifically, the image analyzer 27 convolves the kernel Kx on a group of pixel values arranged in 3 rows and 3 columns centered on the pixel value I[22], and then converts the pixel value I[23] into Spatial filtering processing is performed on all pixel values, in which a kernel Kx is convolved with a central group of pixel values arranged in 3 rows and 3 columns. As a result, image data after spatial filtering processing (that is, image data after edge extraction in the row direction) including pixel values in x rows and y columns is obtained. Further, by performing similar spatial filtering processing using the kernel Ky, image data after column-direction edge extraction can be obtained. Note that the edge extraction filter is not limited to the Sobel filter, and may be various types such as a differential filter and a Prewitt filter.
《Main effects of Embodiment 2》
As described above, by using the method of the second embodiment, the same effects as in the first embodiment can be obtained. Note that whether to use the method of the first embodiment (contrast) or the method of the second embodiment (edge strength) depends on, for example, the types of the inspection object 20a and the background 20b in FIG. 1, and the specifications of the illumination device 16. It may be determined as appropriate depending on the situation. Alternatively, a method may be used in which the determination is made by appropriately combining both contrast and edge strength, instead of using either contrast or edge strength.

(Embodiment 3)
《Details of image analyzer》
FIG. 8 is a schematic diagram showing a configuration example of the image analyzer in FIG. 2 in the image processing apparatus according to Embodiment 3 of the present invention. In the first and second embodiments, contrast and edge strength were used to quantitatively evaluate the illumination condition ILC[i]. However, assuming that the conventional image evaluation criteria by a skilled user is optimal, there is a possibility that a difference will occur between this manual evaluation criteria and the evaluation criteria using contrast or edge strength.

Therefore, in Embodiment 3, AI (Artificial Intelligence) is made to learn the evaluation criteria for images by a skilled user, thereby causing the AI to perform evaluation in place of the skilled user. Specifically, the image analyzer 27 is equipped with AI (Artificial Intelligence) that has been trained to calculate an evaluation value equivalent to the evaluation value based on the user's evaluation criteria. Then, the image analyzer 27 uses AI to calculate an evaluation value for the image data IMD[i] for each illumination condition ILC[i], and outputs the evaluation value as an image analysis result AR[i].

The image analyzer 27 shown in FIG. 8 is equipped with a convolutional neural network (CNN). CNN is widely known as an AI for image recognition. CNN mainly includes an input layer 35, a convolution layer 36, a pooling layer 37, a fully connected layer 38, and an output layer 39. The convolution layer 36 and the pooling layer 37 are not limited to one layer, but may be configured to be stacked alternately.

Image data 31 (IMD) including a plurality of pixel values in the matrix direction is input to the input layer 35 . The convolution layer 36 performs a convolution operation on the image data 31 in the input layer 35 using filters 41a, 41b, and 41c of multiple channels (three channels in this example). Thereby, the convolution layer 36 creates multiple channels of image data 42a, 42b, and 42c. Weighting coefficients W ₀ ¹ , W ₀ ² , and W ₀ ³ are set for the filters 41a, 41b, and 41c, respectively.

The pooling layer 37 performs predetermined arithmetic processing on the plurality of channels of image data 42a, 42b, and 42c created by the convolution layer 36, respectively. As a result, the pooling layer 37 creates multiple channels of image data 43a, 43b, and 43c with a reduced spatial size. Specifically, the pooling layer 37 divides the image data 42a into sections to include a plurality of (2×2 in this example) pixel values, and calculates the maximum value or average value of the plurality of pixel values for each section. The image data 43a is created by calculating the following.

The fully connected layer 38 expands the multiple channels of image data 43a, 43b, and 43c created by the pooling layer 37 into one-dimensional vectors. The output layer 39 outputs an evaluation value EV by performing a sum-of-products operation on each one-dimensional vector in the fully connected layer 38 using a predetermined weighting coefficient Wi.

During learning using such an image analyzer 27, for example, image data 31 and an evaluation value determined by a user who is familiar with the image data 31 are input as teacher data. The image analyzer 27 calculates the error between the evaluation value EV output from the output layer 39 when the image data 31 is used as the input layer 35 and the evaluation value determined by a skilled user. Then, the image analyzer 27 uses the error backpropagation method to bring this error close to zero, and uses the weighting coefficient Wi between the fully connected layer 38 and the output layer 39 and the weighting coefficients of the filters 41a, 41b, and 41c. W ₀ ¹ , W ₀ ² , and W ₀ ³ are updated.

By repeatedly performing such learning, for example, the image features that are important when a skilled user determines the evaluation value are mainly the weighting coefficients W ₀ ¹ , W ₀ ² , W of the filters 41a, 41b, 41c. ₀ ³ . Further, the relationship between the image features and the evaluation value EV is mainly learned as a weighting coefficient Wi. As a result, when the image analyzer 27 supplies predetermined image data 31 to the input layer 35 at the time of inference, the output layer 39 can output an evaluation value EV equivalent to the evaluation value based on the evaluation criteria of a skilled user. become. Furthermore, in this case, the comparator 28 may select the illumination condition that gives the highest evaluation value EV as the final illumination condition ILCx in steps S110 and S111 of FIG.

《Main effects of Embodiment 3》
As described above, by using the method of the third embodiment, the same effects as in the first embodiment can be obtained. Furthermore, depending on how the AI is trained, it becomes possible to further optimize image evaluation criteria when determining lighting conditions. As a result, it becomes possible to improve the inspection accuracy of the inspection system.

Note that the present invention is not limited to the embodiments described above, and includes various modifications. For example, the embodiments described above are described in detail to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to having all the configurations described. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. . Furthermore, it is possible to add, delete, or replace some of the configurations of each embodiment with other configurations.

Further, each of the above-mentioned configurations, functions, processing units, processing means, etc. may be partially or entirely realized in hardware by designing, for example, an integrated circuit. Furthermore, each of the above configurations, functions, etc. may be realized by software by a processor interpreting and executing a program for realizing each function. Information such as programs, tables, files, etc. that implement each function can be stored in a memory, a recording device such as a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

Further, the control lines and information lines are shown to be necessary for explanation purposes, and not all control lines and information lines are necessarily shown in the product. In reality, almost all components may be considered to be interconnected.

10: Inspection system, 15: Imaging device, 16: Lighting device, 17: Lighting control device, 18: Image processing device, 19a to 19c: Cable, 20a: Electronic component (inspection object), 20b: Wiring board (background) , 25: Image interface, 26: Illumination condition searcher, 27: Image analyzer, 28: Comparator, 29: Memory, 31, 31a to 31d: Image data, 35: Input layer, 36: Convolution layer, 37: Pooling layer, 38: fully connected layer, 39: output layer, filters 41a to 41c, 42a to 42c: image data, 43a to 43c: image data, AR: image analysis result, E: edge strength value, EV: evaluation value, I: pixel value, ICS: illumination control signal, ILC: illumination condition, IMD: image data

Claims (12)

An image processing device that processes image data of an inspection target imaged in a state where it is illuminated,
an illumination condition searcher that variably sets the illumination conditions of the illumination;
an image interface that acquires the image data for each of the illumination conditions;
an image analyzer that performs image analysis on the image data for each of the illumination conditions;
a comparator that selects any one of the illumination conditions from the plurality of illumination conditions by comparing image analysis results for each of the illumination conditions by the image analyzer;
has,
Image processing device. The image processing device according to claim 1,
The image analyzer calculates a contrast for the image data for each illumination condition, and outputs the contrast as the image analysis result.
Image processing device. The image processing device according to claim 1,
The image analyzer extracts edges from the image data for each of the illumination conditions, calculates an intensity value of the extracted edge, and outputs the intensity value of the edge as the image analysis result.
Image processing device. The image processing device according to claim 1,
The image analyzer is equipped with AI (Artificial Intelligence) that has been trained to calculate an evaluation value equivalent to the evaluation value based on the user's evaluation criteria, and applies the AI to the image data for each of the lighting conditions. calculating the evaluation value using the method, and outputting the evaluation value as the image analysis result.
Image processing device. A lighting device that illuminates the object to be inspected;
an imaging device that creates image data by imaging the object to be inspected in a state where the illumination is applied;
an image processing device that processes the image data;
An inspection system having
The image processing device includes:
a lighting condition searcher that variably sets lighting conditions of the lighting device;
an image interface that acquires the image data for each of the illumination conditions from the imaging device;
an image analyzer that performs image analysis on the image data for each of the illumination conditions;
a comparator that selects any one of the illumination conditions from the plurality of illumination conditions by comparing image analysis results for each of the illumination conditions by the image analyzer;
has,
Inspection system. The inspection system according to claim 5,
The image analyzer calculates a contrast for the image data for each illumination condition, and outputs the contrast as the image analysis result.
Inspection system. The inspection system according to claim 5,
The image analyzer extracts edges from the image data for each of the illumination conditions, calculates an intensity value of the extracted edge, and outputs the intensity value of the edge as the image analysis result.
Inspection system. The inspection system according to claim 5,
The image analyzer is equipped with AI (Artificial Intelligence) that has been trained to calculate an evaluation value equivalent to the evaluation value based on the user's evaluation criteria, and applies the AI to the image data for each of the lighting conditions. calculating the evaluation value using the method, and outputting the evaluation value as the image analysis result.
Inspection system. A lighting device that illuminates the object to be inspected;
an imaging device that creates image data by imaging the object to be inspected in a state where the illumination is applied;
An inspection method for inspecting the object to be inspected using an inspection system comprising:
a first step of variably setting lighting conditions of the lighting device;
a second step of acquiring the image data for each of the illumination conditions from the imaging device and performing image analysis on the image data for each of the illumination conditions;
a third step of selecting one of the plurality of illumination conditions by comparing image analysis results for each of the illumination conditions;
has,
Inspection method. In the testing method according to claim 9,
In the second step, a contrast is calculated for the image data for each of the illumination conditions, and the contrast is output as the image analysis result.
Inspection method. In the testing method according to claim 9,
In the second step, edges are extracted from the image data for each illumination condition, intensity values of the extracted edges are calculated, and the intensity values of the edges are output as the image analysis results.
Inspection method. In the testing method according to claim 9,
In the second step, the evaluation is performed on the image data for each lighting condition using AI (Artificial Intelligence) that has been trained to calculate an evaluation value equivalent to the evaluation value based on the user's evaluation criteria. A value is calculated, and the evaluation value is output as the image analysis result.
Inspection method.

Images