JP5824278B2 - Image processing device - Google Patents

Description

  The present invention relates to an image processing apparatus used for inspecting an image of an inspection object.

  Conventionally, inspecting an image of an object to be inspected visually by an inspector in the process of manufacturing, assembling, or inspecting an industrial product has a poor inspection accuracy and is not efficient. An image processing apparatus that automatically determines non-defective products is used. As this type of image processing apparatus, there is an apparatus that is connected to a line sensor using a line-shaped solid-state imaging device and captures a one-dimensional image in which pixels are arranged in a line (hereinafter referred to as “line image”).

  This image processing apparatus captures a plurality of line images while moving the inspection object relative to the line sensor, and arranges the captured plurality of line images in time series to form a two-dimensional image in which pixels are arranged two-dimensionally Is generated. This two-dimensional image is used, for example, for inspection of non-defective / defective products of the inspection object.

  By the way, a technique has been proposed in which an area sensor is used to acquire a line image by taking in only specific scanning line data (see, for example, Patent Document 1). If this technology is used, the image processing apparatus can acquire either a two-dimensional image in which pixels are two-dimensionally arranged by a camera or a line image.

JP 2004-108828 A

  However, since the above-described image processing apparatus acquires a line image for one line for each imaging from the camera, for example, in order to perform an inspection using images of a plurality of locations of the inspection target, It is necessary to image a plurality of times, and the inspection time becomes long. Alternatively, the image processing apparatus requires a plurality of cameras (or line sensors) in order to acquire images at a plurality of locations on the inspection object with one imaging.

  The present invention has been made in view of the above-described reasons, and an object thereof is to provide an image processing apparatus capable of reducing the inspection time while acquiring images of a plurality of locations of an inspection object with a single camera. .

The image processing apparatus according to the present invention extracts a preset image in an area image from a camera that uses an area image sensor that can capture an image of an inspection object as an area image in which pixels are two-dimensionally arranged. An image capturing unit that sequentially captures an image of a region as a partial image, in which the relative positional relationship between the inspection object and the camera changes with time, and a storage unit that stores a plurality of the partial images; An image generation unit that generates a two-dimensional image in which pixels are two-dimensionally arranged by arranging the plurality of partial images in the storage unit in time series, and a position and a range of the extraction region in the area image A setting storage unit that stores setting data for determining the partial image capturing condition in the image capturing unit, and the setting data is determined according to an input from an operating device And a control unit for writing the setting storage unit Te, the control unit, said plurality of regions spaced from each other in the moving direction intersecting the direction of the test object within said area image to said extraction region Setting data is determined, the storage unit stores the partial image for each extraction region, and the image generation unit generates the two-dimensional image for each extraction region.

  In this image processing apparatus, it is desirable that the extraction region is in a line shape and the partial image constitutes a line image.

  In this image processing apparatus, the control unit configures the two-dimensional image corresponding to each designated area by designating a plurality of rectangular areas as designated areas by the operating device in the area image. It is desirable to automatically determine the setting data for capturing the partial image.

  The image processing apparatus further includes a display unit that displays the area image captured from the camera as a setting screen on a display device, and the control unit is configured to determine a position and a range of the designated area displayed on the setting screen. More preferably, the designated area is designated by adjusting according to the input from the operating device.

  The present invention has an advantage that the inspection time can be shortened while acquiring images of a plurality of locations of the inspection object with a single camera.

1 is a schematic block diagram illustrating a configuration of an image processing apparatus according to a first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. 6 is an explanatory diagram of a setting screen of the image processing apparatus according to the first embodiment. FIG. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. FIG. 4 is an operation explanatory diagram of the image processing apparatus according to the first embodiment. 1 is a schematic configuration diagram of an inspection system using an image processing apparatus according to a first embodiment. 1 is a layout diagram of an inspection system using an image processing apparatus according to Embodiment 1. FIG. FIG. 2 is an operation explanatory diagram of an inspection system using the image processing apparatus according to the first embodiment. FIG. 9 is an operation explanatory diagram of the image processing apparatus according to the second embodiment.

(Embodiment 1)
The image processing apparatus according to the present embodiment is connected to a camera that captures an image of an inspection object, and is used to determine whether the inspection object is a non-defective product or a defective product.

  As shown in FIG. 1, the image processing apparatus 1 includes a front processing unit 11 connected to the camera 2. The image capturing unit 111 captures an image from the camera 2, and the image capturing unit 111 captures the image. The pre-processing unit 11 includes a storage unit 112 that stores images. Further, the image processing apparatus 1 is connected to a main processing unit 12 that performs image processing on an image captured by the preprocessing unit 11, a setting storage unit 13 that stores various parameters, and a display device (not shown). A display unit 14 that performs output and a control unit 15 that controls each part are provided.

  Here, the image processing apparatus 1 includes a computer having a CPU and a memory. In addition to the camera 2, an operation device (not shown) such as a dedicated keypad and a mouse, and a console 3 having a display device, as will be described later. The external device 4 composed of an encoder or the like can be connected.

As the camera 2, an area image sensor capable of capturing a two-dimensional image (hereinafter referred to as “area image”) in which pixels are two-dimensionally arranged is used as an image sensor. In the present embodiment, a CMOS (Complementary) is included in the image sensor of the camera 2. Metal Oxide Semiconductor (Image Oxide Semiconductor) image sensor is used, but not limited to this, for example, CCD (Charge Coupled) Other solid-state imaging devices such as Devices) may be used. The camera 2 is a color camera that captures a color image, but may be a camera that captures a grayscale image having a grayscale value as a pixel value.

  The camera 2 not only has a function of outputting an area image, but also captures a video signal (pixel value) from only a predetermined range of pixels of the image sensor, thereby extracting an image of an extraction region set in the area image (hereinafter, “ It also has a function of outputting a “partial image”. Here, the extraction region is set in a line shape having a predetermined width, and the partial image constitutes a line image in which pixels are arranged in a straight line with one or a plurality of pixel widths. That is, the camera 2 is a control circuit that can switch between two operation modes: an area imaging mode that outputs an entire normal area image, and a partial imaging mode that outputs a line-shaped partial image that is a part of the area image ( (Not shown).

  The control circuit switches the operation mode according to the signal from the image processing apparatus 1. In the partial imaging mode, the camera 2 performs designation of a pixel for capturing a video signal in accordance with a signal from the image processing apparatus 1. That is, the camera 2 sets an extraction area for capturing an image as a partial image in the area image in accordance with a signal from the image processing apparatus 1. Specifically, in response to an instruction from the control unit 15, the image capturing unit 111 transmits a signal instructing switching of the operation mode (area imaging mode / partial imaging mode) of the camera 2 to the camera 2.

  Since the image processing apparatus 1 of the present embodiment basically operates the camera 2 in the partial imaging mode, hereinafter, it is assumed that the camera 2 operates in the partial imaging mode unless otherwise specified.

  When the control unit 15 of the image processing apparatus 1 receives an instruction to start inspection from the external device 4 or the console 3 connected to the image processing apparatus 1, the control unit 15 instructs the pre-processing unit 11 to acquire an image. When recognizing that an image has been acquired by the preprocessing unit 11, the control unit 15 instructs the main processing unit 12 to perform image processing on the acquired image. The main processing unit 12 extracts and digitizes the image feature amount (for example, the area of the defective portion) according to the inspection parameter, automatically determines the quality of the inspection object, and displays the result on the display unit 14 or the control unit. The data is output from the unit 15 to the external device 4 or the console 3.

  The setting storage unit 13 includes imaging condition parameters used at the time of image acquisition by the front processing unit 11, inspection parameters used for image processing by the main processing unit 12, operation parameters for image processing results by the main processing unit 12, and the like. Information necessary for the operation of the image processing apparatus 1 is stored. The imaging condition parameter here is data indicating the position and range of the extraction region in the area image, and includes at least setting data for determining the partial image capturing condition in the image capturing unit 111.

  In addition to the image capturing unit 111 and the storage unit 112, the preprocessing unit 11 includes a synchronization processing unit 113, an image generation unit 114, an image storage unit 115, and an optimization processing unit 116 as shown in FIG. I have.

  The image capturing unit 111 sequentially captures the images of the extraction region as partial images in a state where the relative positional relationship between the inspection object and the camera 2 changes with time. That is, while moving at least one of the inspection object and the camera 2, the image capturing unit 111 sequentially captures the image of the extraction area as a partial image and writes it in the storage unit 112.

  At this time, the image capturing unit 111 captures a partial image from the camera 2 under a capturing condition determined by setting data in the setting storage unit 13. Specifically, the image capturing unit 111 has a communication function with the camera 2, and according to the setting data in the setting storage unit 13, the camera determines which position and range of the area image is the extraction region. 2 is instructed. Thereby, the camera 2 outputs the image of the area designated as the extraction area to the image processing apparatus 1 as a partial image according to the setting data in the setting storage unit 13. The setting storage unit 13 may be provided in the camera 2.

  The synchronization processing unit 113 generates a trigger signal for determining the timing at which the image capturing unit 111 captures an image (partial image). The synchronization processing unit 113 receives signals from the console 3 and the external device 4, and sends a trigger signal to the image capturing unit 111 so that the image capturing unit 111 can capture a partial image in synchronization with these signals. Is output. Here, the synchronization processing unit 113 generates two types of trigger signals, a head trigger and a partial trigger, and determines the partial image capture timing in the image capture unit 111 by the partial trigger. That is, the image capturing unit 111 transmits the trigger signal (partial trigger) generated by the synchronization processing unit 113 to the camera 2, and receives the trigger signal (partial trigger) to output the partial image from the camera 2. Are stored in the storage unit 112.

  The image generation unit 114 arranges a plurality of partial images stored in the storage unit 112 in time series, and connects the plurality of partial images to generate a two-dimensional image in which pixels are two-dimensionally arranged. . Specifically, the image generation unit 114 monitors the start trigger and the partial trigger generated by the synchronization processing unit 113, and when detecting the start trigger, the image generation unit 114 writes the partial image. Set the address to the initial address.

  Thereafter, when detecting the partial trigger, the image generation unit 114 reads the image data of the partial image stored in the storage unit 112, writes the image data in the image storage unit 115 according to the write address, and sets the write address to the partial image. Shift by the image size of. Then, each time the partial generation is detected, the image generation unit 114 repeatedly executes the writing of the partial image to the image storage unit 115 and the process of shifting the write address, and the partial image is stored in the image storage unit 115. A two-dimensional image is generated in the image storage unit 115 side by side. When the two-dimensional image in the image storage unit 115 reaches a preset image size, the image generation unit 114 outputs a signal to the synchronization processing unit 113 to notify the completion of the generation of the two-dimensional image.

  In the setting storage unit 13, information such as the number of partial images constituting the two-dimensional image, the image size of the partial image, the initial address indicating the writing area of the image storage unit 115, and the size of the two-dimensional image are registered in advance. The image generation unit 114 generates a two-dimensional image using these pieces of information. The number of partial images constituting the two-dimensional image is specified in the setting data in the setting storage unit 13 as the number of captures, and the image generation unit 114 generates a two-dimensional image from the partial images for the number of captures. To do. The synchronization processing unit 113 outputs a leading trigger every time it outputs partial triggers for the number of acquisitions.

  The optimization processing unit 116 performs image processing in the main processing unit 12 by performing various conversion processes such as resizing and noise removal on the partial image captured by the image capturing unit 111 according to the setting. An optimization process for generating a suitable partial image is performed. Although optimization processing is not necessarily required, when the optimization processing unit 116 performs the optimization processing, the image generation unit 114 generates a two-dimensional image using the partial image after the optimization processing. That is, the image generation unit 114 transfers the image data of the partial image buffered in the storage unit 112 to the write address of the image storage unit 115 after the optimization processing unit 116 performs optimization processing. Thus, a two-dimensional image is generated.

  Thus, the pre-processing unit 11 reconstructs a two-dimensional image using the partial image captured from the camera 2. The specific operation of each part of the pre-processing unit 11 (the image capturing unit 111, the synchronization processing unit 113, the image generation unit 114, and the optimization processing unit 116) depends on the imaging condition parameters stored in the setting storage unit 13. It is determined and switched for each inspection condition of the inspection object. Specific operations of the synchronization processing unit 113, the image generation unit 114, and the optimization processing unit 116 will be described later.

  The pre-processing unit 11 performs a shading correction unit (not shown) that performs shading correction on the two-dimensional image generated by the image generation unit 114 so as to suppress luminance unevenness in the longitudinal direction (horizontal direction) of the partial image. ) May be further provided. In short, since the partial image captured by the camera 2 may vary in luminance in the longitudinal direction due to the influence of the characteristics of the image sensor, lens aberration, and the like, the shading correction unit corrects this luminance variation. Uneven brightness can be suppressed.

  By the way, in this embodiment, the extraction area is set at a plurality of locations in one area image, and the image capturing unit 111 captures partial images for all of the plurality of extraction areas and stores them for each extraction area. Store in the unit 112. For this reason, the storage unit 112 stores a plurality of partial images captured with time for each extraction region.

  Further, the image generation unit 114 performs a process of generating a two-dimensional image by arranging a plurality of partial images in time series as described above for each of a plurality of extraction regions set in the area image. Therefore, the image storage unit 115 stores a two-dimensional image reconstructed by arranging the partial images in time series for each extraction region.

  As a result, since a plurality of images (two-dimensional images) are acquired in the pre-processing unit 11, the image processing in the main processing unit 12 is also executed for each two-dimensional image. That is, the main processing unit 12 associates with each extraction region what kind of image processing is performed, and each of the two-dimensional images acquired for each extraction region by the preprocessing unit 11 corresponds to each image. Execute the process.

  Next, as shown in FIG. 2A, in the case where two areas of the first area 101 and the second area 102 are set as extraction areas in the area image 100, the pre-processing unit 11 The operation will be described. In the example of FIG. 2, both the first area 101 and the second area 102 are set along the horizontal line (horizontal scanning line) of the area image 100 at a position near the upper end of the area image 100. The first area 101 and the second area 102 are separated from each other in the horizontal direction in the area image 100.

  In this case, the preprocessing unit 11 determines the timing of image capture in the image capture unit 111 at the synchronization processing unit 113, and the image generation unit 114 uses a plurality of partial images captured at this timing. Reconstruct a two-dimensional image. Here, the inspection object is conveyed toward the upper side of FIG. 2A with respect to the camera 2, and the synchronization processing unit 113 triggers a partial trigger every time the inspection object moves relative to the camera 2. Is generated. Therefore, as shown in FIG. 2B, the pre-processing unit 11 sequentially captures partial images of the two extraction areas of the first area 101 and the second area 102 in synchronization with the movement of the inspection object. Then, the partial image is stored in the storage unit 112 for each extraction region.

  Further, the preprocessing unit 11 distinguishes the first region 101 and the second region 102 by the image generation unit 114 and repeats the writing of the partial image to the image storage unit 115 for each extraction region, thereby extracting the extraction region. A two-dimensional image is generated every time. In short, as shown in FIG. 2C, the pre-processing unit 11 connects the first two-dimensional image 103 obtained by connecting the partial images of the first area 101 and the second image obtained by connecting the partial images of the second area 102. Two two-dimensional images with the two-dimensional image 104 are generated. In FIG. 2C, two-dimensional images 103 and 104 are configured by arranging a plurality of partial images (in the figure, a range separated by a broken line) in the vertical direction (up and down direction in the figure). .

  As described above, the front processing unit 11 generates the two-dimensional images 103 and 104 by arranging a plurality of partial images captured by the image capturing unit 111 in time series for each extraction region while the inspection object moves. To do. As a result, the preprocessing unit 11 can obtain the two-dimensional images 103 and 104 corresponding to the rectangular designated areas 105 and 106 (see FIG. 2A) in the area image 100.

  Here, the image capturing unit 11 captures only a partial image from the camera 2 and requires data transfer from the camera 2 to the image processing apparatus 1 for each imaging as compared to capturing the entire area image. You can save time. In short, for example, when the first region of area Sa and the second region of area Sb are set as the extraction regions for the area image of area S (S> Sa + Sb), the image processing apparatus 1 is only for partial images. By taking in, the time for taking in the image data can be shortened. Specifically, the relationship between the time Ts required for data transfer of the entire area image and the times Ta and Tb required for data transfer of the partial images in the first region and the second region is expressed as “S: Sa + Sb ≈Ts: Ta + Tb ”. Therefore, the image processing apparatus 1 can sample at a high speed by capturing only the partial image, as compared with the case of capturing the entire area image, by the area ratio between the area image and the partial image.

  By the way, the image processing apparatus 1 of the present embodiment allows the user to arbitrarily set the position and range of the extraction region in the area image, that is, the user arbitrarily sets setting data for determining the partial image capturing condition. The following configuration is adopted so that it is possible.

  That is, information on imaging condition parameters, inspection parameters, and operation parameters stored in the setting storage unit 13 is input by a user using a console (operation device) 3 connected to the image processing apparatus 1, and the control unit 15 is Via the setting storage unit 13. In other words, the control unit 15 has a function of determining setting data and the like according to an input from the operation device of the console 3 and writing the setting data in the setting storage unit 13.

  In the present embodiment, as described above, since the extraction areas are set at a plurality of locations in one area image, the control unit 15 sets the setting data so that a plurality of areas separated from each other in the area image are set as the extraction areas. To decide.

  Here, the image processing apparatus 1 has two operation modes of an inspection mode for capturing an image (partial image) from the camera 2 and a setting mode in which setting data is set. The operation mode is switched accordingly.

  In the setting mode, the control unit 15 instructs the image capturing unit 111 to switch the operation mode of the camera 2 to the area imaging mode, and switches the operation mode of the camera 2 to the area imaging mode. In this state, the control unit 15 controls the display unit 14 to display the area image of the inspection object captured from the camera 2 by the front processing unit 11 on the display device of the console 3 as a setting screen. At this time, the display unit 14 displays the area being selected as the extraction area on the setting screen on the setting screen. In other words, the display device displays the area image as the setting screen 107 as shown in FIG. 3, and the first region 101 and the second region 102 are displayed on the setting screen 107 as candidates for the extraction region (shaded portion in the figure). indicate.

  Further, the control unit 15 adjusts the positions and ranges of the first area 101 and the second area 102 on the setting screen 107 in accordance with the input from the console 3. That is, the control unit 15 receives a user operation on the console 3 and adjusts the coordinate position and the vertical and horizontal width dimensions of the first area 101 and the second area 102 on the setting screen 107 to Adjust the position and range of the extraction area. At this time, the control unit 15 blinks, for example, the area on the setting target side of the first area 101 and the second area 102 and sets the area to be set according to the operation of the console 3 by the user. Switch.

  The control unit 15 registers the position and range of the extraction region (the first region 101 and the second region 102) adjusted in this way in the setting storage unit 13 as setting data. Thereby, the position of the extraction area (first area 101, second area 102) adjusted on the setting screen 107 is reflected in the setting data in the setting storage unit 13.

  Here, in this embodiment, there are a plurality of types (a plurality of types) of inspection objects, and the setting storage unit 13 stores setting data (extraction region position and range) for each type of inspection object. In the inspection mode, the pre-processing unit 11 specifies the type of the inspection object in accordance with the input from the console 3, and reads the setting data corresponding to this type from the setting storage unit 13, thereby changing the type of the inspection object. Apply the appropriate setting data.

  Hereinafter, a method for the user to set the setting data in the image processing apparatus 1 configured as described above will be briefly described.

  First, the user operates the console 3 to switch the operation mode of the image processing apparatus 1 to the setting mode. Thereby, for example, a setting screen 107 as shown in FIG. 3 is displayed on the display device. In this state, the user operates the console 3 to individually move the first area 101 and the second area 102 on the setting screen 107 and adjust the position of the extraction area. When the position of the extraction region is determined, the user operates the operation device 3 to set the range of the extraction region, and then the control unit 15 sets the adjusted positions and ranges of the first region 101 and the second region 102. The data is registered in the setting storage unit 13 as data.

  According to the image processing apparatus 1 of the present embodiment described above, the control unit 15 determines the setting data so that a plurality of areas separated from each other in the area image are the extraction areas. A plurality of partial images in the image can be acquired from the camera 2. Furthermore, since the storage unit 112 stores partial images for each extraction region, and the image generation unit 114 generates a two-dimensional image for each extraction region, the main processing unit 12 performs image processing corresponding to each extraction region. Can be executed for each extraction region. Accordingly, the image processing apparatus 1 can acquire images at a plurality of locations on the inspection object with one camera 2 while acquiring images at a plurality of locations on the inspection target, thereby reducing the inspection time. Is possible.

  Further, since the control unit 15 determines setting data indicating the position and range of the extraction region in accordance with an input from the operation device (console 3), the partial image is not limited to a line image having a width of one pixel, but a plurality of pixels. It may be a rectangular image having a width of. That is, the extraction region is not limited to a line shape, and may be a rectangular shape, for example. Even when the relative position between the camera 2 and the inspection object continuously changes due to the partial image having a certain width, the image processing apparatus 1 captures the partial image without omission in the entire inspection object. Thus, the entire inspection object can be inspected.

  By the way, the control unit 15 captures a partial image constituting a two-dimensional image corresponding to each designated area by designating a plurality of rectangular areas as designated areas in the operation device of the console 3 in the area image. The setting data may be automatically determined. In this case, instead of directly specifying the position and range of the extraction region, the control unit 15 is required to construct the two-dimensional image by specifying the region from which the two-dimensional image is to be acquired as the specified region. The position and range of the extraction region for taking in a partial image are determined.

  For example, in the area image 100 of FIG. 2A, when the designated area 105 corresponding to the left half is designated and the designated area 106 corresponding to the right half is designated, the control unit 15 specifies these designated areas 105 and 106. The setting data is determined so that the two-dimensional image can be reconstructed. That is, the control unit 15 extracts the first area 101 for acquiring a partial image constituting one designated area 105 and the second area 102 for obtaining a partial image constituting the other designated area 106. The setting data is determined so as to be an area. Thereby, the pre-processing unit 11 can finally obtain the two-dimensional images 103 and 104 corresponding to the rectangular designated areas 105 and 106 in the area image 100.

  When determining the setting data based on the rectangular designated area in this way, the control unit 15 displays the rectangular designated area on the setting screen in place of the area selected as the extraction area. 14 is controlled. In this case, the control unit 15 determines the designated area by adjusting the position and range of the designated area displayed on the setting screen according to the input from the console 3.

  According to this configuration, the user can specify a desired two-dimensional image as an inspection target by designating the acquisition range of the two-dimensional image as the designated region. That is, the user can indirectly determine the position and range of the extraction region by designating the designated region without directly determining the position and range of the extraction region, and setting data can be easily set. There is an advantage.

  Hereinafter, specific operations of the synchronization processing unit 113, the image generation unit 114, and the optimization processing unit 116 will be described.

  The operation of the synchronization processing unit 113 differs depending on the form of the inspection object, the conveyance speed of the inspection object, and the like.

  First, a case will be described in which an inspection object that does not have a separation between individual inspection objects and requires continuous inspection, such as a fluid or a band, is described. In this case, as illustrated in FIG. 4, the image capturing unit 111 acquires an image of the inspection object 5 from the camera 2 during a period between the start signal and the end signal given from the external device 4 or the console 3. .

  Here, when the inspection object is transported at a constant speed, the synchronization processing unit 113 generates a partial trigger for determining the timing for capturing a partial image at a constant time interval T as shown in FIG. To do. The image capturing unit 111 detects the rising edge of the partial trigger and outputs a trigger signal (partial trigger) to the camera 2. When the camera 2 receives the trigger signal from the image capturing unit 111, a predetermined exposure time is obtained. The imaging is performed over the period. Here, the time interval T of the partial trigger is set in the imaging condition parameter of the setting storage unit 13 so as to be longer than at least the time t required to capture the partial image. Thereby, the image capturing unit 111 can capture the partial image without missing.

  Further, when the inspection object conveyance speed is accelerated or decelerated, based on the output of an encoder that is provided in a conveyance device (not shown) that conveys the inspection object and generates a pulse signal that is synchronized with the inspection object conveyance speed. The synchronization processing unit 113 generates a partial trigger. That is, the synchronization processing unit 113 counts the pulses output from the encoder as the external device 4 and generates a partial trigger every time the pulse count value reaches the specified value C as shown in FIG. . Here, the specified value C is set such that the period CT at which the count value reaches the specified value C when the conveyance speed is maximum is longer than at least the time t required to capture the partial image.

  In the case of an inspection object that requires continuous inspection, the number of partial images to be captured is, for example, the number of partial images captured while the inspection object moves by a unit length, or the number of partial images captured per unit time. By setting the number of partial images to be checked, it becomes easy to manage information on the inspection object.

  In addition, since the synchronization processing unit 113 captures a continuous inspection object without missing, for the second and subsequent head triggers, the generation processing of the two-dimensional image by the image generation unit 114 is monitored, and the generation of the two-dimensional image is completed. The first trigger is generated at the timing when is detected. Accordingly, the synchronization processing unit 113 can perform continuous inspection without inputting another signal from the outside of the image processing apparatus 1.

  Upon receiving an end signal from the external device 4 or the console 3, the preprocessing unit 11 stops generating a trigger by the synchronization processing unit 113 and ends capturing an image (partial image) from the camera 2.

  Next, a description will be given of the case of inspecting discontinuous inspection objects that are independent from each other, such as panels and electronic components that are transported randomly. In this case, a sensor (not shown) for detecting the inspection object is provided in the transport device, and the synchronization processing unit 113 uses a signal from this sensor as an external trigger signal as shown in FIG. At the same time, the top trigger is generated. In addition, when the movement of the inspection object is synchronized with the output of the encoder of the transport apparatus, and one inspection object is imaged for each rotation of the encoder, the synchronization processing unit 113 uses the encoder instead of the sensor. The inspection object may be detected by using a Z-phase pulse generated every rotation.

  However, when the synchronization processing unit 113 directly uses the external trigger signal from the sensor as the head trigger, the capture start timing of the partial image constituting the two-dimensional image is fixed depending on the installation position of the sensor. Accordingly, the image generation unit 114 may generate a two-dimensional image in which the entire inspection object 5 is not included and a part of the inspection object 5 is missing, as illustrated in FIG. 7A, for example. is there.

  In order to avoid such a problem, the synchronization processing unit 113 has an offset function for generating a timing for generating the head trigger with a delay time from the timing at which the signal from the sensor is detected. The offset amount (delay time) is set to a certain time when the inspection object is being conveyed at a constant speed, and is set by the number of pulses of the encoder when the conveyance speed of the inspection object is being accelerated / decelerated. Is done. With this offset function, the pre-processing unit 11 can arbitrarily adjust the head position of the two-dimensional image and generate a two-dimensional image including the entire inspection object 5 as shown in FIG. it can.

  Note that the operation in which the synchronization processing unit 113 generates a partial trigger is the same as that in the case of an inspection object that requires continuous inspection, even in the case of an independent inspection object. That is, when the inspection object is conveyed at a constant speed, the synchronization processing unit 113 generates a partial trigger at a constant time interval from the head trigger. Further, when the conveyance speed of the inspection object is accelerated or decelerated, the synchronization processing unit 113 generates a partial trigger based on the output of the encoder of the conveyance device.

  In the case of an inspection object that requires continuous inspection, the image generation unit 114 may cause a defect (feature point) to be divided in the generated two-dimensional image depending on the timing of generating the two-dimensional image. There is sex. For example, as shown in FIG. 8, when the dirt 50 on the surface of the strip-shaped inspection object 5 is detected as a defect, the dirt 50 is divided into two two-dimensional images 103, so that each two-dimensional image 103 The size of the detected dirt 50 is about half the size of the original dirt 50.

  Therefore, in order to avoid this problem, when the image generation unit 114 updates the frame in response to the head trigger, the last part of the two-dimensional image generated immediately before (the last partial image captured) is updated. Copy to the beginning of the dimensional image. As a result, as shown in FIG. 9, one of the two two-dimensional images 103 generated in succession always includes a defect (here, “dirt 50”) without being divided, and the main process The part 12 can determine the exact size of the defect.

  The optimization processing unit 116 can execute a plurality of types of optimization processing, and performs the optimization processing according to the inspection parameters set according to the inspection condition of the inspection object. In the present embodiment, the optimization processing unit 116 can execute three types of processing as reduction processing (compression processing), enlargement processing, and division processing as optimization processing.

  The reduction process is a process for obtaining an average value, a median value, a maximum value, and a minimum value of pixel values (luminance values) in a preset m × n size block for an M × N size partial image. Yes, it is a process of reducing the size of the partial image by rearranging the obtained values. That is, the optimization processing unit 116 removes a noise component in the conveyance direction of the inspection object by, for example, taking the median luminance value for n pixels arranged in the inspection object conveyance direction with respect to the partial image. Can do. In this case, the size of the partial image in the conveyance direction of the inspection object is reduced to 1 / n.

  The enlargement process is a process for obtaining a new pixel value for each pixel constituting the partial image using the luminance values of the pixels around the target pixel. By rearranging the obtained pixel values, the size of the partial image is determined. It is a process to enlarge. That is, for example, as shown in FIG. 10A, the optimization processing unit 116 obtains a new pixel value for the partial image of the Bayer array captured from the color camera using the luminance values of the pixels around the target pixel. Thus, the RGB data shown in FIG. 10B can be converted. In this case, the partial image is enlarged three times in size.

  The division processing is performed by performing some filtering process on the partial image, so that one partial image is subjected to a plurality of partial images that have been subjected to different filtering processes, or a partial image that has not been subjected to the filtering process and a partial image that has been subjected to the filtering process. It is a process which produces | generates. That is, the optimization processing unit 116 performs a differentiation process on the partial image using a Sobel filter, for example, and divides the partial image into two images, the differential image and the original partial image. The filter processing here is filter processing in an image processing technique, and includes a differential filter that extracts a differential value (change amount) of a pixel value, a directional differential filter that extracts a directional differential value, and the like.

  Next, an inspection system that inspects an inspection object using the image processing apparatus 1 of the present embodiment will be described. Here, an inspection system is exemplified in which the inspection object is a panel such as a building board, and the inspection items include the presence or absence of scratches (irregularities) on the panel surface and the presence or absence of white turbidity or uneven color in the panel.

  As shown in FIG. 11, the inspection system includes, in addition to the image processing apparatus 1 and the camera 2, a transport apparatus 6 that transports the inspection object 5 and an illumination apparatus 7 that illuminates the inspection object 5 from above the transport apparatus 6. I have.

  The conveyance device 6 is composed of, for example, a roller conveyor, and is driven by an induction motor 61 to convey the inspection object 5 placed on the conveyance surface 60 (see FIG. 12) in one direction (right direction in FIG. 11). The transport device 6 includes a sensor 62 that is disposed above the transport surface 60 and detects the inspection object 5, and an encoder 63 that is attached to the rotating shaft of the roller and detects the amount of movement of the inspection object 5. . The conveying device 6 is connected to the image processing device 1 so that the output signals of the sensor 62 and the encoder 63 are input to the image processing device 1 (the connection lines are not shown in FIG. 11).

  In the transport device 6, the moving speed of the inspection object 5 is adjusted within a range of the maximum speed Vmax (m / min) or less by control of an inverter (not shown) that supplies power to the induction motor 61. The encoder 63 of the transport device 6 outputs a P (pulse) signal with respect to the movement amount L (mm) of the inspection object 5.

  In this inspection system, as shown in FIG. 12, the camera 2 is fixed above the transport surface 60 of the transport device 6 so that the optical axis is parallel to the normal line of the transport surface 60. As a result, the camera 2 takes a fixed area centered directly below the camera 2 on the transport surface 60 of the transport device 6 as an inspection area, and captures an image of this inspection area as an area image. In the area image captured in this way, the upstream side of the transport device 6 is the upper side, and the upper left corner is the origin of the camera coordinates, that is, the coordinate position (H, V) = (0, 0) (H is horizontal) The coordinate value of the axis, V is the coordinate value of the vertical axis).

  The illumination device 7 includes a first illumination 71 and a second illumination 72 each having a linear (line-shaped) light emitting surface extended in a direction orthogonal to the conveyance direction within a plane along the conveyance surface 60. ing. The first illumination 71 is disposed downstream of the camera 2 in the transport direction and at a position lower than the camera 2 from the transport surface 60. The 1st illumination 71 irradiates diffused light toward the conveyance surface 60 so that the downstream half of a conveyance direction may mainly be illuminated among the visual fields (inspection area | region) of the camera 2. FIG.

  On the other hand, the second illumination 72 is disposed substantially directly below the camera 2 and at a position where the height from the conveyance surface 60 is lower than that of the first illumination 71. The second illumination 72 is tilted with respect to the transport surface 60 so that the angle θ formed by the optical axis and the transport surface 60 is an acute angle, and mainly upstream in the transport direction in the field of view (inspection region) of the camera 2. Is irradiated with light having a relatively high directivity toward the transport surface 60.

  According to the above configuration, the image (mainly the lower half of the area image) captured by the camera 2 by receiving the light emitted from the first illumination 71 and reflected by the inspection object 5 is the cloudiness of the inspection object 5. The image has a high contrast against uneven color. On the other hand, an image (mainly the upper half of the area image) captured by the camera 2 by receiving light irradiated from the second illumination 72 and reflected by the inspection object 5 is a scratch (unevenness) on the surface of the inspection object 5. On the other hand, the image has a high contrast. That is, when the camera 2 images the inspection object 5 in the area imaging mode, the upper half has high contrast with respect to the scratch on the surface of the inspection object 5, and the lower half has contrast with respect to white turbidity and color unevenness of the inspection object 5. A high area image can be obtained.

  The area image obtained in this way is an image having gradation in the vertical direction (conveying direction), and the contrast with respect to the above-described defects (surface scratches, white turbidity, uneven color) is the highest in a part of the vertical direction. Therefore, in order to stably inspect the inspection object 5 with high accuracy, the image processing apparatus 1 sets, as the extraction region, a region having the highest contrast with respect to the defect in the area image 100 as illustrated in FIG. Specifically, the second region 102 is set on a horizontal line having the highest contrast with respect to a defect (a scratch on the surface of the inspection object 5) in the upper half of the area image 100. The first area 101 is set on a horizontal line having the highest contrast with respect to a defect (white turbidity or uneven color) in the lower half of the area image 100.

  The image generation unit 114 generates a two-dimensional image for inspecting the scratch on the surface of the inspection object 5 from the partial image of the second region 102, and white turbidity and uneven color from the partial image of the first region 101. A two-dimensional image for performing the inspection is generated. The main processing unit 12 can inspect the presence or absence of defects (surface scratches, white turbidity, color unevenness) of the inspection object 5 using these two-dimensional images.

  Here, the synchronization processing unit 113 counts the pulses output from the encoder 63, and captures the partial images of the first region 101 and the second region 102 when the pulse count value reaches the specified value C. Generate a partial trigger. Note that the specified value C times the period (expressed by the maximum speed Vmax, the movement amount L of the inspection object 5 and the number of pulses P) in which the encoder 63 generates one pulse when the conveyance speed is maximum is one part. The specified value C is set so as to be longer than the time required to capture an image. In the present embodiment, the prescribed value C is set according to the resolution of the camera 2 so that the pixel aspect ratio in the real space of one pixel of the partial image is 1: 1.

  With respect to the partial image of the first region 101 acquired in this way, the optimization processing unit 116 performs no processing so that the size of a defect (white turbidity or color unevenness) can be measured. On the other hand, with respect to the partial image of the second region 102, the optimization processing unit 116 sets the maximum luminance value in the conveyance direction (vertical direction) of the inspection object so that the continuity of the defects (surface scratches) becomes clear. Perform the required reduction process.

  The image generation unit 114 includes a two-dimensional image formed by combining a plurality of partial images in the first area 101, and a two-dimensional image formed by combining a plurality of partial images in the second area 102 after optimization processing (reduction processing). Is generated. The main processing unit 12 executes corresponding image processing for each two-dimensional image acquired for each of the extraction regions.

  When the camera 2 is in the area imaging mode, the image capturing unit 111 receives an image acquisition instruction from the control unit 15, outputs a trigger signal to the camera 2, receives the trigger signal, and outputs from the camera 2. The stored image data is stored in the storage unit 112. The image generation unit 114 transfers the image data buffered in the storage unit 112 to the write address of the image storage unit 115, and generates an area image. When the camera 2 is a Bayer array color camera, the image generation unit 114 converts the buffered Bayer array image data into RGB data by the optimization processing unit 116, and then sets the write address in the image storage unit 115. The image is sequentially transferred to generate an area image.

  In the present embodiment, the case where two areas of the first area and the second area are set as extraction areas in the area image has been described as an example. However, the present invention is not limited to this example. Three or more regions may be set as the extraction region.

(Embodiment 2)
The image processing apparatus 1 according to the present embodiment differs from the image processing apparatus 1 according to the first embodiment in that the optimization processing unit 116 functions as a combination processing unit that combines partial images of different extraction regions to generate a combined image. Is different. Hereinafter, the same configurations as those of the first embodiment are denoted by common reference numerals, and the description thereof is omitted as appropriate.

  That is, in the first embodiment, the image generation unit 114 generates a two-dimensional image by arranging a plurality of partial images for each extraction region. However, in the present embodiment, the image generation unit 114 generates a two-dimensional image. Before being generated, partial images of different extraction regions are combined.

  The optimization processing unit 116 according to the present embodiment executes a combination process for generating a combined image by combining partial images of different extraction regions stored in the storage unit 112 as the optimization process. Thereby, the partial images obtained by imaging the parts of the inspection object that are separated from each other are integrated into one image (combined image).

  For example, when the image processing apparatus 1 is used for measuring the width dimension of the inspection object 5 (the horizontal dimension in FIG. 14A) as shown in FIG. 14, the minimum necessary for measuring the width dimension is used. Are captured by the image capturing unit 111 as partial images. That is, in order to measure the width dimension of the inspection object 5, the partial image of the first area 101 including one end edge in the width direction of the inspection object 5 and the second area 102 including the other edge in the width direction are measured. Since partial images are required, the image capturing unit 111 captures these two partial images.

  Here, the camera 2 is arranged such that the width direction of the inspection object 5 is the horizontal direction of the area image 100 (the left-right direction in FIG. 14A), and the entire length of the inspection object 5 is within the area image 100. Direction and field of view are set. In the example of FIG. 14, the first area 101 is set in a range of a plurality of pixels from the left end of the area image 100, and the second area 102 is set in a range of a plurality of pixels from the right end of the area image.

  The partial images 108 and 109 in the first area 101 and the second area 102 are set on the same horizontal scanning line (line related to horizontal scanning) in the area image, as shown in FIG. The area image 100 has a width of one pixel in the vertical direction. The lengths of the first area 101 and the second area 102 are set such that the sum of both is smaller than the horizontal dimension of the area image 100, so that the first area 101 and the second area 102 are on the area image 100. Are set apart from each other in the horizontal direction. Further, since the extraction area (the first area 101 and the second area 102) is set at a fixed position in the area image 100, the interval W0 between the first area 101 and the second area 102 in the horizontal direction is known.

  The optimization processing unit 116 arranges the partial image 108 of the first area 101 and the partial image 109 of the second area 102 captured from the camera 2 in the horizontal direction without gaps, as shown in FIG. The partial images 108 and 109 are combined into one to obtain a combined image. The combined image obtained in this way is an image representing the state of both ends of the inspection object 5 in the width direction.

  The main processing unit 12 measures the dimension W1 from one end edge to the other end edge in the width direction of the inspection object 5 in the combined image, and uses the dimension W1 between the first region 101 and the second region 102 in the area image 100. By adding to the interval W0 (known), the width dimension of the inspection object 5 can be measured. In this case, the image processing apparatus 1 can measure the width dimension by taking in only the partial image necessary for the measurement of the width dimension of the inspection object 5, so that the processing speed of the measurement process can be improved.

  In the present embodiment, the image generation unit 114 generates a two-dimensional image by arranging a plurality of combined images subjected to the combination processing by the optimization processing unit 116 in time series.

  According to the image processing apparatus 1 of the present embodiment described above, a minimum image necessary for image processing is taken as a partial image, and partial images of different extraction regions are combined to generate a combined image. It becomes easy to compare the images of the portions necessary for image processing.

  Other configurations and functions are the same as those of the first embodiment.

1 image processing device 2 camera 3 console (operation device, display device)
111 Image Acquisition Unit 112 Storage Unit 114 Image Generation Unit 115 Image Storage Unit 116 Optimization Processing Unit (Combination Processing Unit)
13 Setting Storage Unit 14 Display Unit 15 Control Unit

Claims (4)

From a camera that uses an area image sensor that can capture an image of an inspection object as an area image in which pixels are arranged two-dimensionally as an image sensor, an image of an extraction region that is set in advance in the area image is used as a partial image. An image capturing unit that sequentially captures the relative positional relationship between the inspection object and the camera as time passes, a storage unit that stores a plurality of the partial images, and a plurality of the parts in the storage unit An image generation unit that generates a two-dimensional image in which pixels are two-dimensionally arranged by arranging images in time series, and data indicating the position and range of the extraction region in the area image, the image capture A setting storage unit that stores setting data for determining the capture condition of the partial image in the capture unit; and the setting data that is determined according to an input from an operating device and written in the setting storage unit And a non-control unit,
The control unit determines the setting data so that a plurality of regions separated from each other in a direction intersecting a moving direction of the inspection target in the area image is the extraction region, and the storage unit The partial image is stored for each region, and the image generation unit generates the two-dimensional image for each extraction region.   The image processing apparatus according to claim 1, wherein the extraction area has a line shape, and the partial image forms a line image.   The control unit captures the partial image constituting the two-dimensional image corresponding to each designated area by designating a plurality of rectangular areas as designated areas in the area image by the operating device. The image processing apparatus according to claim 2, wherein the setting data is automatically determined. A display unit that displays the area image captured from the camera as a setting screen on a display device;
The said control part designates the said designation | designated area | region by adjusting the position and range of the said designation | designated area | region displayed on the said setting screen according to the input from the said operating device. image processing equipment.