JP2016103787A - Image processing device, image processing system, image processing method, and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device which detects dirt and provides a notification about it before the dirt adversely affects an image to be taken.SOLUTION: An image processing device 10 comprises: a generating unit for generating the frequency distribution of the luminance values of plural pixels acquired from a camera; an extracting unit for extracting a luminance value having a highest frequency on the basis of the frequency distribution; a determination unit for determining whether the luminance value extracted by the extracting unit is greater than a predetermined value; and an output unit which outputs a notification when the determination unit determines that the luminance value extracted by the extractor is greater than the predetermined value.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image processing apparatus, an image processing system, an image processing method, and an image processing program.

  An image monitoring apparatus that detects dirt on a monitoring camera from a change in the edge amount of an optical image in a monitoring target area is disclosed (Patent Document 1).

  An image processing apparatus that determines the life of an illumination light source from the brightness difference between an image captured at night when the illumination light source is turned on and an image captured when the illumination light source is turned off is disclosed. (Patent Document 2).

JP 2003-189294 A JP 2005-72767 A

  However, the contamination of the surveillance camera can be detected by the image monitoring apparatus of Patent Document 1 only after the contamination has progressed. Further, the image processing apparatus of Patent Document 2 can determine the life of the illumination light source, but cannot detect contamination of the camera.

  An object of one aspect is to provide an image processing apparatus or the like that can detect dirt before adversely affecting a captured image.

  In one aspect, the image processing apparatus creates a frequency distribution of luminance values of a plurality of pixels acquired from a camera, an extraction unit that extracts a luminance value having the maximum frequency based on the frequency distribution, A determination unit that determines whether or not a luminance value extracted by the extraction unit is greater than a predetermined value; and a determination that the luminance value extracted by the extraction unit is greater than the predetermined value by the determination unit An output unit that outputs a notification.

  In one aspect, an image processing apparatus or the like that can detect dirt before adversely affecting a captured image can be provided.

It is explanatory drawing which shows the example of installation of an image processing apparatus. It is an apparatus block diagram which shows the hardware constitutions of an image processing system. It is explanatory drawing which shows the example of imaging | photography of an image processing apparatus. It is explanatory drawing which shows the example of imaging | photography when an image processing apparatus is dirty. It is explanatory drawing which shows the luminance value frequency distribution figure of an imaging example. It is explanatory drawing which shows the luminance value frequency distribution figure of an imaging example. It is explanatory drawing which shows the luminance value frequency distribution figure of an imaging example. It is explanatory drawing which shows the luminance value frequency distribution figure of an imaging example. It is explanatory drawing which shows the luminance value frequency distribution figure of the imaging example of the vehicle from which a color differs. It is a flowchart which shows the process sequence of an image processing apparatus. 10 is a flowchart illustrating a processing procedure of the image processing apparatus according to the second embodiment. 10 is a flowchart illustrating a processing procedure of the image processing apparatus according to the third embodiment. 15 is a flowchart illustrating a processing procedure of the image processing apparatus according to the fourth embodiment. 15 is a flowchart illustrating a processing procedure of the image processing apparatus according to the fourth embodiment. 10 is a flowchart illustrating a processing procedure of the image processing apparatus according to the fifth embodiment. FIG. 16 is a device configuration diagram illustrating a hardware configuration of a sixth embodiment. 15 is a flowchart illustrating a processing procedure of the image processing apparatus according to the sixth embodiment. 18 is a flowchart illustrating a processing procedure of the image processing apparatus according to the seventh embodiment. 20 is a flowchart illustrating a processing procedure of the image processing apparatus according to the eighth embodiment. FIG. 20 is a functional configuration diagram illustrating an overall configuration of an image processing system according to a ninth embodiment. FIG. 20 is an apparatus configuration diagram illustrating a hardware configuration of an image processing system according to a tenth embodiment.

[Embodiment 1]
FIG. 1 is an explanatory diagram illustrating an installation example of the image processing apparatus 10. An image processing apparatus 10 according to the present embodiment is an apparatus that captures a license plate of a vehicle that is passing, reads characters written on the license plate from an image, and outputs them. The image processing apparatus 10 according to the present embodiment estimates the degree of contamination of the imaging window 15 provided on the surface of the image processing apparatus 10 from the frequency distribution or average value of the luminance values of the captured image, and the contamination is license plate. This is notified to other devices before it adversely affects the accuracy of reading the characters described in.

  The image processing apparatus 10 is attached to a traffic light installed beside a road. The image processing apparatus 10 is attached so that the photographing window 15 and the illumination window 16 face the road.

  FIG. 2 is an apparatus configuration diagram illustrating a hardware configuration of the image processing system. The image processing system includes an image processing apparatus 10, a server 30, a network 31, and a communication line 41.

  The image processing apparatus 10 includes an illumination device 11, a camera 33, a communication module 14, a photographing window 15, an illumination window 16, a CPU (Central Processing Unit) 20, a memory 18, and a clock 19. The camera 33 includes the photographing optical system 12 and the image sensor 13.

  The illuminating device 11 includes an infrared LED (Light Emitting Diode), and irradiates the imaging target region with infrared rays at night. The lighting device 11 is turned on and off by a clock 19 or a brightness sensor (not shown).

  The imaging optical system 12 is a combination of optical components such as a lens, a diaphragm, and an optical filter, and forms an optical image of a subject on an imaging surface provided on the surface of the imaging element 13. The photographing optical system 12 includes an infrared cut filter (not shown). The infrared cut filter is inserted between the subject and the image sensor 13 in the daytime to shield unnecessary infrared rays and enables clear photographing with visible light. On the other hand, the infrared cut filter is removed from between the subject and the image sensor 13 at night. Therefore, the infrared reflected light irradiated by the illumination device 11 enters the image sensor 13.

  A plurality of photoelectric conversion elements 131 are two-dimensionally arranged on the imaging surface of the imaging element 13. The image sensor 13 photoelectrically converts the optical image formed by the photographing optical system 12 by each photoelectric conversion element 131 and outputs an electric signal in a predetermined format. In the following description, an electric signal having a predetermined format is referred to as image data, and an operation in which the image sensor 13 outputs image data is referred to as photographing. Further, an image data output in a form that can be recognized by a human being on a display device such as a liquid crystal display or a medium such as paper is called an image. The image processing apparatus 10 of the present embodiment does not output an image during normal use, but will be described below using an image for convenience. In the following description, it is assumed that the so-called black-and-white type image sensor 13 having no color filter is used.

  The CPU 20 is an arithmetic control device that processes image data captured by the image sensor 13 and outputs an image analysis result, and controls the image sensor 13 and peripheral components. As the CPU 20, an image processing semiconductor designed exclusively for image processing, a general-purpose DSP (Digital Signal Processor), one or a plurality of CPUs, a multi-core CPU, or the like is used.

  The memory 18 is a storage device such as SRAM (Static RAM), DRAM (Dynamic RAM), flash memory, or semiconductor memory disk. The memory 18 stores information necessary during the processing performed by the CPU 20 and a program executed by the CPU 20.

  The clock 19 outputs the time. The clock 19 is preferably capable of adjusting the time to a standard time such as Japan Standard Time via the communication module 14.

  The communication module 14 performs communication between the image processing apparatus 10 and the network 31 via the communication line 41. The communication module 14 is used when outputting the description of the license plate of the vehicle to be passed and the notification regarding the contamination of the image processing apparatus 10 to the server 30 via the network 31. The communication module 14 and the network 31 are connected by wire or wireless. A large number of image processing apparatuses 10 are connected to the network 31.

  The server 30 includes a mass storage device 42, a communication module 43, a CPU 44, and a memory 45. The server 30 is a general-purpose computer, for example. The communication module 43 performs communication between the server 30 and the network 31 via the communication line 41. The CPU 44 is an arithmetic and control unit that processes the data received by the communication module 43 and stores it in the mass storage device 42. Information stored in the large-capacity storage device 42 is displayed on a display device (not shown), or is output to another terminal such as a computer or a smartphone via the communication module 43.

  The memory 45 is a storage device such as SRAM, DRAM, flash memory, or semiconductor memory disk. The memory 18 stores information necessary during the processing performed by the CPU 44 and a program executed by the CPU 44.

 The output destination from the image processing apparatus 10 via the network 31 may be various electronic devices such as a smartphone and a tablet terminal.

  The image processing apparatus 10 is housed in a housing (not shown). A photographing window 15 that transmits visible light and infrared light is provided in the vicinity of the photographing optical system 12 of the housing. An illumination window 16 that transmits infrared rays is provided in the vicinity of the illumination device 11 of the housing. The photographing optical system 12 may be exposed outside the housing without providing the photographing window 15. In this case, in order to prevent dust and the like from entering the inside of the housing, it is desirable that the gap be not generated between the outer surface of the photographing optical system 12 and the housing, for example.

  FIG. 3 is an explanatory diagram illustrating a shooting example of the image processing apparatus 10. The license plate on the front side of the running vehicle is photographed. FIG. 4 is an explanatory diagram illustrating an example of photographing when the image processing apparatus 10 is dirty. If dirt such as dust adheres to the photographing window 15, light is scattered. For this reason, the image is taken as if bright portions of light such as the white license plate and the periphery of the vehicle light are blurred. This makes it difficult to read the characters written on the license plate. As the stain progresses further, the entire image becomes whitish and the characters written on the license plate cannot be read.

  5 to 8 are explanatory diagrams showing luminance value frequency distribution diagrams of the photographing example. The horizontal axis represents the luminance value, and the vertical axis represents the number of pixels. In the present embodiment, one photoelectric conversion element 131 corresponds to one pixel. Since the monochrome image sensor 13 is used, the luminance value is the output of each photoelectric conversion element 131. In this embodiment, since the imaging device 13 with an 8-bit output is used, the luminance value is an integer from 0 to 255. Pixels with a luminance value of 0 are black and pixels with a luminance value of 255 are white.

  FIG. 5 shows a luminance value frequency distribution diagram of a photographing example when the photographing window 15 is not dirty. In FIG. 5, the average luminance value is 109. It can be seen that there is a large mountain in the range where the luminance value is about 70 to 100, and there are many pixels indicating the luminance value in this range. There is a mountain with a luminance value as small as about 230. This is the effect of particularly bright parts such as headlights. The recognition rate when the photographing window 15 is not dirty is 97%. Here, the recognition rate means the ratio of images that the CPU 20 can read all the characters displayed on the license plate out of the images taken by the image processing apparatus 10. The 3% image including unreadable characters includes an image having a situation caused by the subject such as a dirty license plate.

  FIG. 6 shows a luminance value frequency distribution diagram of a photographing example when the photographing window 15 starts to become dirty. In FIG. 6, the average luminance value is 121. In FIG. 6, there are large peaks in the range where the luminance value is about 90 to 120, and small peaks where the luminance value is about 230. Compared to FIG. 5, the large mountain is shifted to the right, and the heights of the large mountain and the small mountain are approaching. The recognition rate when the shooting window 15 is dirty in FIG. 6 is 97%. In the stage of FIG. 6, the recognition rate is equivalent to the case without dirt.

  FIG. 7 shows a luminance value frequency distribution diagram of a photographing example when the photographing window 15 is further dirty. In FIG. 7, the average luminance value is 142. In FIG. 7, there are peaks in the range where the luminance value is about 130 to 140 and the luminance value is about 230. Compared with FIG. 5 and FIG. 6, a peak having a luminance value of about 230 rises sharply, and the heights of the two peaks are reversed. The recognition rate when the shooting window 15 is dirty in FIG. 7 is 95%. In the stage of FIG. 7, the recognition rate is slightly lower than when there is no dirt, but is within an allowable range for the operation of the image processing apparatus 10.

  FIG. 8 is a luminance value frequency distribution diagram of a photographing example when the photographing window 15 is further dirty than FIG. In FIG. 8, the average luminance value is 165. In FIG. 8, there is a smooth mountain with a luminance value of about 130 to 160, and a sharp mountain with a luminance value of about 230. Compared with FIGS. 5 and 6, the heights of the two peaks are reversed. The recognition rate when the photographing window 15 is dirty in FIG. 8 is 42%. In the stage of FIG. 8, the recognition rate is too low to be acceptable in operation.

  As described above, when the stage of FIG. 7 is detected and the image processing apparatus 10 is cleaned, preventive maintenance that maintains a good recognition rate can be performed while suppressing the frequency of cleaning.

  When the image processing apparatus 10 is not cleaned, the time required for the photographing window 15 to become dirty from the state of FIG. 5 to the state of FIG. 8 varies greatly depending on the installation location of the image processing apparatus 10 and the weather. For example, roads along the coast exposed to sea breeze get dirty in as little as three months. On the other hand, there is a place with the degree of contamination as shown in FIG. Furthermore, the photographing window 15 is also displayed due to various changes in circumstances such as a construction where dust is generated in the vicinity of the installation location, a change in the surrounding environment such as the felling of trees, or a change in the amount of traffic of the vehicle. The time required to get dirty from the state of 5 to the state of FIG. 8 varies.

  FIG. 9 is an explanatory diagram illustrating a luminance value frequency distribution diagram of a photographing example of a vehicle having a different color. FIG. 9 is an example of photographing when the photographing window 15 is not dirty. FIG. 9A shows a case where a white vehicle is photographed. In the luminance value frequency distribution diagram, the luminance value D is the maximum value S. FIG. 9B shows a case where a blue vehicle is photographed. In the luminance value frequency distribution diagram, the luminance value E is the maximum value T. FIG. 9C shows a case where a black vehicle is photographed. In the luminance value frequency distribution diagram, the luminance value F is the maximum value U. 9A is on the right side of FIGS. 9B and 9C. 9C is on the left side of FIGS. 9A and 9B. In this way, the luminance value frequency distribution diagram has different shapes depending on the color and shape of the subject vehicle. Hereinafter, the processing procedure of the image processing apparatus 10 will be described using the luminance value frequency distribution diagram of FIG. 9 as an example.

  FIG. 10 is a flowchart showing a processing procedure of the image processing apparatus 10. The flow of processing performed by the CPU 20 of the present embodiment will be described using FIG.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). Here, the object to be processed is image data including a vehicle license plate as shown in FIGS. Specifically, for example, the CPU 20 performs image recognition processing on the acquired image data, and determines whether or not a square plate having a predetermined aspect ratio is included. If it is included, the CPU 20 determines that the image data is an object to be processed. The image sensor 13 may repeat shooting continuously at a fixed timing, or only when the vehicle passes an appropriate position in cooperation with a vehicle sensor (not shown) installed on the road. Also good.

  If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image. In the case of image data to be processed (YES in step S401), the CPU 20 creates a frequency distribution of luminance values of the image data (step S402). The frequency distribution is created by counting the pixels having the luminance value for each luminance value. It should be noted that the pixels may be counted for each luminance value, or the pixels may be counted by dividing the luminance value into two or three divisions.

  The CPU 20 extracts the luminance value that maximizes the frequency based on the frequency distribution (step S403). In the following description, this luminance value is referred to as the maximum frequency luminance value. For example, when the frequency distribution of the image being processed is FIG. 9A, the luminance value D is the luminance value E when the frequency distribution is FIG. 9B, and when the frequency distribution is the case of FIG. It is.

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). The predetermined number may be determined by the number of processed image data such as 200, or may be determined by time such as 1 hour. If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of processes have been completed (YES in step S404), the CPU 20 obtains the average value of the luminance values at the maximum frequency extracted in step S403 (step S405).

  The CPU 20 determines whether or not the average value obtained in step S405 is larger than a predetermined threshold value (step S406). The threshold value is indicated by TH on the horizontal axis of FIGS. 9A to 9C, for example. When the image processing apparatus 10 is not dirty or has little dirt, the luminance values D, E, and F at which the frequency distribution is maximum are smaller than the threshold value TH as shown in FIG. Accordingly, NO is determined in step S406. On the other hand, when the image processing apparatus 10 is dirty, as described with reference to FIGS. 7 and 8, the luminance value at which the frequency distribution is maximum is larger than the threshold value TH. Therefore, it is determined YES in step S406.

  If it is determined that the average value is equal to or less than the threshold value TH (NO in step S406), the CPU 20 ends the process. If it is determined that the average value is larger than the threshold value TH (YES in step S406), the CPU 20 outputs a notification that the image processing apparatus 10 is dirty (step S407). Thereafter, the process ends. The notification is output by the communication module 14. The output notification is sent to the server 30 via the network 31.

  According to the present embodiment, as described with reference to FIG. 7, it is possible to realize the image processing apparatus 10 that issues a notification when the dirt is accumulated but the recognition rate is acceptable.

  The image processing apparatus 10 may be housed in a single housing, or two or more of the lighting device 11 and the camera 33 and the circuit portion including the CPU 20, the communication module 14, the memory 18, and the clock 19. It may be stored separately in a number of cases. Further, only the photographing optical system 12 may be separated from other parts. If the image processing apparatus 10 is housed in one housing, it is easy to mount and maintain. When divided into the illumination device 11 and the camera 33 and the circuit portion, the camera 33 having specifications such as an appropriate number of pixels, sensitivity, and angle of view can be selected and used according to the installation location. Further, when the photographing optical system 12 is separated, the photographing optical system 12 having an appropriate angle of view and focal length can be selected and used according to the installation location.

  For example, so-called pixel addition may be performed, in which outputs of a plurality of photoelectric conversion elements 131 such as four are combined and handled as one pixel. In this way, since the sensitivity of the image sensor 13 is increased, the description on the license plate can be read without reducing the output of the illumination device 11 or without using the illumination device 11.

  A color-type imaging device 13 having a color filter on the subject side of the photoelectric conversion device 131 may be used. In this case, in step S402, for example, a frequency distribution can be created using only the output of the photoelectric conversion element 131 having a green color filter. Further, after performing a so-called demosaicing process for complementing colors, a frequency distribution may be created using luminance values calculated by weighting for each color. If it does in this way, in addition to the character described in the license plate in the daytime, the color of a vehicle can also be determined, and the character described in the license plate can be determined even at low illumination at night.

  The image processing apparatus 10 may perform the process illustrated in FIG. 10 only when the illumination device 11 is lit. The phenomenon described with reference to FIGS. 4 to 9 appears particularly clearly when the reflected light of the light from the illumination device 11 is photographed using the camera 33 adjacent to the illumination device 11. 15 stains can be accurately detected.

  The image processing apparatus 10 may be used for so-called machine vision applications such as an image inspection apparatus in various production lines or a visual sensor of an industrial robot. In any case, it is possible to realize the image processing apparatus 10 that issues a notification before the deterioration of the image quality due to dirt affects the function of the apparatus. The notification may be output by turning on a marker lamp attached in the vicinity of the image processing apparatus 10.

[Embodiment 2]
The present embodiment relates to an image processing apparatus 10 that estimates the degree of contamination of a photographing window 15 from the average value of luminance values of photographed images. FIG. 11 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the second embodiment. The process flow of the present embodiment will be described with reference to FIG. Note that description of portions common to the first embodiment is omitted.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image.

  In the case of image data to be processed (YES in step S401), the CPU 20 calculates the average luminance value of the image data shown in FIG. 9 (step S411).

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of processes have been completed (YES in step S404), the CPU 20 obtains an average value of the average luminance values calculated in step S411 (step S412).

  The CPU 20 determines whether or not the average value is larger than a predetermined threshold value (step S406). As described with reference to FIGS. 5 to 8, when the image processing apparatus 10 becomes dirty, the average luminance value increases. As shown in FIG. 7, it is desirable to set an average luminance value as a threshold value that starts to become dirty but does not cause a large change in the recognition rate.

  If it is determined that the average value is smaller than the threshold value (NO in step S406), the CPU 20 ends the process. If it is determined that the average value is greater than the threshold (YES in step S406), the CPU 20 outputs a notification that the image processing apparatus 10 is dirty (step S407). Thereafter, the process ends. Note that the communication module 14 outputs the notification. The output notification is sent to the server 30 via the network 31.

  In step S411, an arithmetic average, a geometric average, a harmonic average, or the like can be used as the average luminance value. Further, an intermediate value of luminance values may be used instead of the average value.

  According to the present embodiment, it is possible to realize an image processing apparatus 10 that issues a notification at a time when dirt is accumulated but the recognition rate is acceptable, using a simple calculation method.

[Embodiment 3]
The present embodiment relates to an image processing apparatus 10 that estimates the degree of contamination of a photographing window 15 from both the frequency distribution and average value of luminance values of a photographed image. FIG. 12 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the third embodiment. The process flow of the present embodiment will be described with reference to FIG. Note that description of portions common to Embodiment 1 or Embodiment 2 is omitted.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image.

  In the case of image data to be processed (YES in step S401), the CPU 20 calculates the average luminance value of the image data shown in FIG. 9 (step S411). The CPU 20 creates a frequency distribution of luminance values of the image data (step S402). The CPU 20 extracts the luminance value at which the frequency is maximum based on the frequency distribution, that is, the luminance value at the maximum frequency (step S403).

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of processes have been completed (YES in step S404), the CPU 20 obtains the average luminance value calculated in step S411 and the average value of the maximum frequency luminance value detected in step S403, respectively (step S415).

  The CPU 20 determines whether or not the average value of the average luminance value and the average value of the luminance value at the maximum frequency are each greater than a predetermined threshold value (step S417).

  When it is determined that the average value of either one or both is equal to or less than the threshold value (NO in step S417), the CPU 20 ends the process. When it is determined that both average values are larger than the threshold (YES in step S417), the CPU 20 outputs a notification that the image processing apparatus 10 is dirty (step S407). Thereafter, the process ends. Note that the communication module 14 outputs the notification. The output notification is sent to the server 30 via the network 31.

  According to the present embodiment, since the two methods are determined together, it is possible to realize the image processing apparatus 10 that issues a notification regarding dirt with higher accuracy.

[Embodiment 4]
The present embodiment relates to an image processing apparatus 10 that outputs a preliminary notification before cleaning is required. 13 and 14 are flowcharts showing the processing procedure of the image processing apparatus 10 according to the fourth embodiment. The processing flow of the present embodiment will be described with reference to FIGS. 13 and 14. Note that description of portions common to Embodiment 3 is omitted.

  The processing flow of the present embodiment is the same as the processing flow of the third embodiment shown in FIG. 12 up to step S415.

  The CPU 20 determines whether or not the average value of the average luminance value and the average value of the luminance value at the maximum frequency are each equal to or less than a predetermined threshold value (step S451). If it is determined that both average values are equal to or less than the threshold value (YES in step S451), the CPU 20 ends the process.

  If it is determined that one of the average values is larger than the threshold value (NO in step S451), the CPU 20 proceeds to step S417.

  The CPU 20 determines whether or not the average value of the average luminance value and the average value of the luminance value at the maximum frequency are larger than a predetermined threshold value (step S417).

  If it is determined that one of the average values is equal to or less than the threshold (NO in step S417), the CPU 20 outputs a first notification indicating that the image processing apparatus 10 has started to become dirty (step 452). Thereafter, the process ends. If it is determined that both average values are greater than the threshold (YES in step S417), the CPU 20 outputs a second notification indicating that the image processing apparatus 10 is dirty (step S453). Thereafter, the process ends. Note that the communication module 14 outputs the first notification or the second notification. The output first notification or second notification is sent to the server 30 via the network 31.

  According to the present embodiment, it is possible to realize the image processing apparatus 10 that outputs a preliminary notification when it starts to become dirty. When cleaning the image processing apparatus 10 that has output the second notification, if there is an image processing apparatus 10 that has output the first notification in the vicinity of the image processing apparatus 10, it is installed in a wide range by performing cleaning together. The image processing apparatus 10 can be efficiently cleaned.

[Embodiment 5]
In the present embodiment, the frequency distribution of the brightness value of the photographed image is divided into a low-to-medium brightness region and a high-luminance region, and the maximum value of the frequency distribution in each region is compared to estimate the degree of contamination of the photographing window 15. The present invention relates to the image processing apparatus 10 that performs the processing. FIG. 15 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the fifth embodiment. The processing flow of this embodiment will be described with reference to FIGS. 15 and 9. Note that description of portions common to the first embodiment is omitted.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image.

  In the case of image data to be processed (YES in step S401), the CPU 20 creates a frequency distribution of luminance values of the image data (step S402). The CPU 20 extracts the maximum frequency in an area below a predetermined threshold TH, that is, in a low to medium luminance area (step S421). For example, in the frequency distribution diagram shown in FIG. 9A, in step S421, S is extracted as the maximum frequency in the low to medium luminance region. Similarly, in the case of FIG. 9B, the maximum frequency in the low to medium luminance region is extracted as T, and in the case of FIG. 9C, U is extracted as the maximum frequency in the low to medium luminance region.

  The CPU 20 extracts the maximum frequency in the region where the luminance value is equal to or higher than the threshold value TH, that is, in the high luminance region (step S423). For example, in the case of the frequency distribution chart shown in FIG. 9A, P is extracted as the maximum frequency in the high luminance region in step S423. Similarly, in the case of FIG. 9B, the maximum frequency in the high luminance region is extracted as Q, and in the case of FIG. 9C, R is extracted as the maximum frequency in the high luminance region.

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of sheets has been processed (YES in step S404), the CPU 20 obtains the average value of the maximum frequencies extracted in step S421 and the average value of the maximum frequencies extracted in step S423, respectively (step S425).

  The CPU 20 compares the average value of the maximum frequencies of the low to medium luminance regions with the average value of the maximum frequencies of the high luminance regions (step S427). If it is determined that the average value of the maximum frequency values in the low to medium luminance areas is equal to or greater than the average value of the maximum frequency values in the high luminance area (NO in step S427), the CPU 20 ends the process. When it is determined that the average value of the maximum frequencies of the low to medium brightness areas is less than the average value of the maximum frequencies of the high brightness areas (YES in step S427), the CPU 20 outputs a notification to that effect (step S407). . The CPU 20 then ends the process. Note that the communication module 14 outputs the notification. The output notification is sent to the server 30 via the network 31.

  According to the present embodiment, it is possible to realize the image processing apparatus 10 that gives notification at a time when dirt is accumulated but the recognition rate is acceptable. Note that, similarly to the third or fourth embodiment, the determination may be performed in combination with the average value of the luminance values.

[Embodiment 6]
The present embodiment relates to the image processing apparatus 10 that automatically cleans the photographing window 15 when it is determined that it is dirty, and outputs a notification only when the cleaning effect is not sufficiently obtained.

  FIG. 16 is a device configuration diagram illustrating a hardware configuration of the sixth embodiment. FIG. 17 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the sixth embodiment. This embodiment will be described with reference to FIGS. 16 and 17. Note that description of portions common to the first embodiment is omitted.

  The image processing apparatus 10 includes a cleaning device 17. The cleaning device 17 is a heater that heats the photographing window 15 to remove dew condensation, frozen ice, or attached snow. The operation and stop of the cleaning device 17 are controlled by a temperature sensor or a humidity sensor (not shown).

  The processing flow of the present embodiment is the same as the processing flow of the first embodiment shown in FIG. 10 up to step S406. If it is determined that the average value is equal to or less than the threshold value (NO in step S406), the CPU 20 ends the process. If it is determined that the average value is greater than the threshold (YES in step S406), the CPU 20 determines whether or not the photographing window 15 has been cleaned (step S431). Here, “cleaned” means that the cleaning device 17 is activated and the imaging window 15 is heated continuously for a predetermined time or more. If the photographing window 15 has not been cleaned (NO in step S431), the CPU 20 operates the cleaning device 17 to start cleaning (step S433). Thereafter, the CPU 20 returns to step S401.

  When the photographing window 15 has been cleaned (YES in step S431), the CPU 20 outputs a notification that the image processing apparatus 10 is dirty (step S407). The CPU 20 then ends the process. Note that the communication module 14 outputs the notification. The output notification is sent to the server 30 via the network 31.

  According to the present embodiment, since the notification is issued only when there is dirt that cannot be removed by the photographing apparatus 10, the photographing apparatus 10 can be efficiently cleaned.

  In addition, when using the illuminating device 11 with a small calorific value, it is efficient that the cleaning device 17 heats the illuminating window 16 simultaneously with the photographing window 15. The cleaning device 17 may be a wiper that wipes the outer surface of the photographing window 15.

[Embodiment 7]
The present embodiment relates to an image processing apparatus 10 that obtains one frequency distribution from a plurality of photographed image data and estimates the degree of contamination of the photographing window 15. FIG. 18 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the seventh embodiment. The processing flow of this embodiment will be described with reference to FIG. Note that description of portions common to Embodiment 1 or Embodiment 2 is omitted.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image. In the case of image data to be processed (YES in step S401), the CPU 20 stores the image data in the memory 18 (step S461).

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of processes have been completed (YES in step S404), the CPU 20 creates a frequency distribution of the luminance values of the predetermined number of image data stored in the memory 18 in step S461 (step S463). The frequency distribution in step S463 is preferably created at a time when the amount of traffic of the vehicle is relatively small and the CPU 20 has a sufficient load.

  Based on the frequency distribution, the CPU 20 extracts the luminance value at which the frequency is maximum, that is, the luminance value at the maximum frequency (step S465). The CPU 20 determines whether or not the maximum frequency luminance value is larger than a predetermined threshold value (step S467).

  When it is determined that the maximum frequency luminance value is equal to or less than the threshold (NO in step S467), the CPU 20 ends the process. If it is determined that the maximum brightness value is greater than the threshold value (YES in step S467), the CPU 20 outputs that effect (step S407). Thereafter, the process ends. Note that the communication module 14 outputs the notification. The output notification is sent to the server 30 via the network 31.

  According to the present embodiment, since the frequency distribution of a plurality of images is created collectively, it is suitable for the image processing apparatus 10 installed in a place where the traffic volume is uneven.

[Embodiment 8]
The present embodiment relates to an image processing apparatus 10 that can automatically set a threshold for determining whether or not there is dirt.

  FIG. 19 is a flowchart illustrating a processing procedure of the image processing apparatus 10 according to the eighth embodiment. The present embodiment will be described with reference to FIG. Note that description of portions common to the first embodiment is omitted.

  The present embodiment starts immediately after the image processing apparatus 10 is attached or immediately after maintenance is performed, that is, the image processing apparatus 10 is not dirty.

  The CPU 20 acquires image data from the image sensor 13 and determines whether or not it is a target to be processed (step S401). If it is not the image data to be processed (NO in step S401), the CPU 20 returns to step S401 and waits for the input of the next image.

  In the case of image data to be processed (YES in step S401), the CPU 20 reads characters in the number plate being photographed (step S441). The CPU 20 determines whether all the characters on the license plate have been read (step S442).

  If there is a character that cannot be read in the license plate (NO in step S442), the CPU 20 proceeds to step S404. If all the characters in the license plate have been read (YES in step S442), the CPU 20 creates a frequency distribution of the luminance values of the image data (step S402). The CPU 20 extracts the luminance value at which the frequency is maximum based on the frequency distribution, that is, the luminance value at the maximum frequency (step S403).

  The CPU 20 determines whether or not a predetermined number of processes have been completed (step S404). Here, it is desirable that the predetermined number is, for example, the number of vehicles traveling overnight or one week at night.

  If the predetermined number of processes has not been completed (NO in step S404), the CPU 20 returns to step S401. If the predetermined number of processes has been completed (YES in step S404), the CPU 20 obtains the license plate unreadable rate (step S443). The unreadable rate is a ratio of license plates including characters that could not be read among all license plates read by the CPU 20 in step S441.

  The CPU 20 compares the unreadable rate with a predetermined threshold value (step S445). If the unreadable rate is equal to or greater than the threshold (NO in step S445), the CPU 20 issues a notification to that effect (step S447). This is because if the unreadable rate is equal to or greater than the threshold value even though the image processing apparatus 10 is not dirty, it is necessary to correct the mounting position or mounting direction of the image processing apparatus 10. Thereafter, the CPU 20 ends the process.

  If the unreadable rate is less than the threshold value (YES in step S445), the CPU 20 sets and stores a threshold value for detecting that the image processing apparatus 10 is dirty based on the statistical value of the luminance value of the image data (step S44). After that, the process is terminated. The threshold value is, for example, a value obtained by adding three times the standard deviation to the average value of the luminance values at the maximum frequency detected in step S403. Alternatively, the maximum value of the luminance value at the maximum frequency detected in step S403 is set as the threshold value. Alternatively, a threshold value is set by adding or integrating a predetermined numerical value to the average value of luminance values at the maximum frequency detected in step S403.

  According to the present embodiment, it is possible to determine a threshold value for detecting contamination of the image processing apparatus 10 in accordance with the place where the image processing apparatus 10 is installed and the characteristics of the vehicle passing through the place.

  Note that after notifying in step S447 that the unreadable rate is greater than or equal to the threshold value, the process may proceed to step S449 to set the threshold value of the maximum brightness value. This threshold is a provisional threshold until the attachment position or orientation of the image processing apparatus 10 is corrected.

  The calculation of the unreadable rate in step S443 may use a ratio between the number of characters that have been read and the number of characters that could be read.

  For example, when the number of unreadable characters is two or less, the determination in step S442 may proceed to step S402 to create a frequency distribution. In this way, the threshold can be set even in places where there are many vehicles that are difficult to read due to dirty dirt plates due to circumstances such as an unpaved road nearby.

  Note that the threshold value of the average value of the average luminance values described in the second embodiment may be set by using a method similar to that described in the present embodiment.

[Embodiment 9]
Embodiment 9 relates to an embodiment showing a configuration of functional blocks of an image processing system. FIG. 20 is a functional configuration diagram illustrating the overall configuration of the image processing system according to the ninth embodiment. The image processing system according to the present embodiment includes an image processing apparatus 10 and a server 30 that are connected to each other via a communication line 41.

  The image processing apparatus 10 realizes the functions of an acquisition unit 21, a statistics unit 22, a determination unit 23, a reading unit 26, an output unit 27, and an identification unit 28. In the image processing apparatus 10 according to the present embodiment, the CPU 20 executes the image processing program stored in the memory 18 so that each function of FIG. 20 is realized.

  The acquisition unit 21 acquires image data output from the image sensor 13 via the bus. The reading unit 26 and the statistical unit 22 perform processing independently on the captured image data.

  The reading unit 26 processes the image data acquired by the acquisition unit 21 and reads characters written on the license plate. If there is a character that cannot be read in the license plate, the reading unit 26 outputs that character as “unknown”. The reading unit 26 reads the number plate for all vehicles that pass there, except when there is a situation such as maintenance.

  The statistics unit 22 processes the image data acquired by the acquisition unit 21 and calculates the statistical value of the luminance value of each pixel constituting the image. The statistics unit 22 may operate constantly, or may operate only for a limited time such as 1 hour per day.

  The statistics unit 22 includes a creation unit 29, an extraction unit 24, and an average unit 25. The creation unit 29 creates a frequency distribution of luminance values of each pixel constituting the image. The extraction unit 24 represents information on the vertical axis or horizontal axis of the mountain portion when the frequency distribution is represented by a frequency distribution diagram with the luminance value of the pixel on the horizontal axis and the number of pixels indicating each luminance value on the vertical axis. To extract. The average unit 25 calculates the average value of the luminance values of each pixel.

  The determination unit 23 compares the output of the extraction unit 24 or the average unit 25 with a predetermined threshold value to determine whether or not a notification needs to be output.

  The output unit 27 outputs a notification based on the determination result of the determination unit 23. The output unit 27 outputs the license plate characters read by the reading unit 21. The output unit 27 is the communication module 14 connected to the CPU 20 via a bus.

  The identification unit 28 is connected to a clock 19 or a brightness sensor (not shown), and identifies whether it is daytime or nighttime. Depending on the information from the identification unit 28, the operation of the statistics unit 22 can be started or stopped.

  The server 30 realizes the function of the reception unit 40. In the server 30 of the present embodiment, the CPU 44 executes the server program stored in the memory 45, thereby realizing the function of the accepting unit.

  The accepting unit 40 accepts data output from the network 31 via the communication line 41. The receiving unit 40 stores the received data in a predetermined format in the mass storage device 42.

[Embodiment 10]
The tenth embodiment relates to a form in which the general-purpose computer 39, the image processing program 32, the general-purpose camera 33, and the general-purpose infrared illumination device 11 are operated in combination to realize the image processing apparatus 10. FIG. 21 is a device configuration diagram illustrating a hardware configuration of the image processing system according to the tenth embodiment. The configuration of this embodiment will be described with reference to FIG. Note that description of portions common to the first embodiment is omitted.

  The image processing apparatus 10 according to the present embodiment uses a general-purpose computer 39 including a CPU 20, a disk drive 34, a communication module 14, a memory 18, a clock 19, and a bus.

  The image processing program 32 is recorded on a portable recording medium 35. The CPU 20 reads the image processing program 32 via the disk drive 34 and stores it in the memory 18. The CPU 20 may read out the image processing imaging program 32 stored in the semiconductor memory 36 such as a flash memory mounted in the computer 39. Further, the CPU 20 may download the image processing program 32 from another server computer (not shown) connected via the communication module 14 and the network 31 and store it in the memory 18.

  The computer 39 reads the image processing program 32 for executing the above-described various software processes from the portable recording medium 35 or the semiconductor memory 36 or downloads it from another server computer (not shown) via the network 31. The image processing program 32 is installed as a control program for a general-purpose computer 39, loaded into a main storage device (not shown), and executed. Thereby, the computer 39 controls the camera 33 and the illumination device 11 and functions as the image processing apparatus 10 described above as a whole.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
A determination unit that determines whether the luminance value extracted by the extraction unit is greater than a predetermined value;
An image processing apparatus comprising: an output unit that outputs a notification when the determination unit determines that the luminance value extracted by the extraction unit is greater than the predetermined value.

(Appendix 2)
An average unit for obtaining an average value of luminance values of a plurality of pixels acquired from the camera;
A determination unit for determining whether the average value is larger than a predetermined value;
An image processing apparatus comprising: an output unit that outputs a notification when the determination unit determines that the average value is larger than the predetermined value.

(Appendix 3)
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
An average unit for obtaining an average value of the acquired luminance values;
A first determination unit that determines whether the luminance value extracted by the extraction unit is greater than a first predetermined value;
A second determination unit for determining whether the average value is greater than a second predetermined value;
The luminance value extracted by the extraction unit by the first determination unit is determined to be larger than the first predetermined value, and the average value is larger than the second predetermined value by the second determination unit. An image processing apparatus comprising: an output unit that outputs a notification when it is determined.

(Appendix 4)
The output unit is
When the first determination unit determines that the luminance value extracted by the extraction unit is greater than the first predetermined value, or the second determination unit determines that the average value is greater than the second predetermined value When it is determined to be large, the first notification is output,
The luminance value extracted by the extraction unit by the first determination unit is determined to be larger than the first predetermined value, and the average value is larger than the second predetermined value by the second determination unit. The image processing apparatus according to appendix 3, wherein a second notification is output when it is determined that.

(Appendix 5)
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
A high-luminance extraction unit that extracts a maximum frequency in a range of luminance values equal to or higher than a predetermined value based on the frequency distribution;
A low luminance side extraction unit that extracts a maximum frequency in a range of luminance values less than the predetermined value based on the frequency distribution;
A determination unit that determines whether or not a value extracted by the high luminance side extraction unit is larger than a value extracted by the low luminance side extraction unit;
And an output unit that outputs a notification when it is determined that the value extracted by the high luminance side extraction unit by the determination unit is larger than the value extracted by the low luminance side extraction unit. apparatus.

(Appendix 6)
The image according to any one of Supplementary Note 1, Supplementary Note 2, and Supplementary Note 5, wherein the determination unit makes a determination based on the luminance value acquired a predetermined number of times from the camera that performs a plurality of times of photographing. Processing equipment.

(Appendix 7)
It has an identification part that identifies that it is nighttime,
The image processing apparatus according to any one of Supplementary Note 1, Supplementary Note 2, Supplementary Note 5, or Supplementary Note 6, wherein the determination unit makes a determination based on the luminance value acquired at night.

(Appendix 8)
An illumination device that irradiates infrared rays to the field of view taken by the camera;
The image processing apparatus according to any one of appendix 1 to appendix 7, wherein the camera is an infrared camera.

(Appendix 9)
The image processing apparatus according to any one of appendices 1 to 8, further comprising a cleaning device that removes surface deposits when the luminance value satisfies a predetermined condition.

(Appendix 10)
Analyzing the image acquired from the camera and having a reading unit that reads characters contained in the subject,
The image processing apparatus according to any one of Supplementary Note 1 to Supplementary Note 9, wherein the output unit outputs characters read by the reading unit.

(Appendix 11)
APPENDIX 1 to APPENDIX 10 comprising: a statistical unit that obtains a statistical value of luminance values of a plurality of pixels acquired a predetermined number of times from the camera; and a setting unit that sets the predetermined value based on the statistical value. The image processing apparatus according to any one of the above.

(Appendix 12)
In an image processing system including an image processing apparatus and a server connected to the image processing apparatus via a communication line,
The image processing apparatus includes:
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
A determination unit that determines whether or not the luminance value extracted by the extraction unit is greater than a predetermined value, and a notification when the determination unit determines that the luminance value extracted by the extraction unit is greater than the predetermined value And an output unit for outputting
The server
An image processing system comprising: a reception unit that receives the notification via the communication line.

(Appendix 13)
In an image processing system including an image processing device and a server connected to the image processing device via a communication line,
The image processing apparatus includes:
An average unit for obtaining an average value of luminance values of a plurality of pixels acquired from the camera;
A determination unit for determining whether the average value is larger than a predetermined value;
An output unit that outputs a notification when the determination unit determines that the average value is greater than the predetermined value;
The server
An image processing system comprising: a reception unit that receives the notification via the communication line.

(Appendix 14)
In an image processing system including an image processing device and a server connected to the image processing device via a communication line,
The image processing apparatus includes:
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
An average unit for obtaining an average value of the acquired luminance values;
A first determination unit that determines whether the luminance value extracted by the extraction unit is greater than a first predetermined value;
A second determination unit for determining whether the average value is greater than a second predetermined value;
The luminance value extracted by the extraction unit by the first determination unit is determined to be larger than the first predetermined value, and the average value is larger than the second predetermined value by the second determination unit. An output unit that outputs a notification when it is determined that
The server
An image processing system comprising: a reception unit that receives the notification via a communication line.

(Appendix 15)
In an image processing system including an image processing device and a server connected to the image processing device via a communication line,
The image processing apparatus includes:
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
A high-luminance extraction unit that extracts a maximum frequency in a range of luminance values equal to or higher than a predetermined value based on the frequency distribution;
A low-luminance side extraction unit that extracts a maximum frequency in a range of luminance values less than the predetermined value based on the frequency distribution;
A determination unit that determines whether or not a value extracted by the high luminance side extraction unit is larger than a value extracted by the low luminance side extraction unit;
An output unit that outputs a notification when the determination unit determines that the value extracted by the high luminance side extraction unit is larger than the value extracted by the low luminance side extraction unit;
The server
An image processing system comprising: a reception unit that receives the notification via a communication line.

(Appendix 16)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Determining whether the extracted luminance value is greater than a predetermined value;
An image processing method comprising: outputting a notification when it is determined that the extracted luminance value is greater than the predetermined value.

(Appendix 17)
Find the average of the brightness values of multiple pixels acquired from the camera,
Determining whether the average value is greater than a predetermined value;
A notification is output when it is determined that the average value is larger than the predetermined value.

(Appendix 18)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Obtain the average value of the acquired luminance values,
Determining whether the extracted luminance value is greater than a first predetermined value;
Determining whether the average value is greater than a second predetermined value;
An image processing method comprising: outputting a notification when it is determined that the extracted luminance value is greater than the first predetermined value and the average value is greater than the second predetermined value.

(Appendix 19)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the maximum frequency in the range of luminance values above a predetermined value,
Extracting a maximum frequency in a range of luminance values less than the predetermined value based on the frequency distribution;
Determining whether a value extracted in a range of luminance values greater than or equal to the predetermined value is greater than a value extracted in a range of luminance values less than the predetermined value;
An image processing method comprising: outputting a notification when it is determined that a value extracted in a range of luminance values greater than or equal to the predetermined value is greater than a value extracted in a range less than the predetermined value.

(Appendix 20)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Determining whether the extracted luminance value is greater than a predetermined value;
An image processing program for causing a computer to execute a process of outputting a notification when it is determined that the extracted luminance value is larger than the predetermined value.

(Appendix 21)
Find the average of the brightness values of multiple pixels acquired from the camera,
Determining whether the average value is greater than a predetermined value;
An image processing program for causing a computer to execute a process of outputting a notification when it is determined that the average value is larger than the predetermined value.

(Appendix 22)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Obtain the average value of the acquired brightness values,
Determining whether the extracted brightness value is greater than a first predetermined value;
Determining whether the average value is greater than a second predetermined value;
An image processing program for causing a computer to execute a process of outputting a notification when it is determined that the extracted luminance value is larger than the first predetermined value and the average value is larger than the second predetermined value.

(Appendix 23)
Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the maximum frequency in the range of luminance values above a predetermined value,
Extracting a maximum frequency in a range of luminance values less than the predetermined value based on the frequency distribution;
Determining whether a value extracted in a range of luminance values greater than or equal to the predetermined value is greater than a value extracted in a range of luminance values less than the predetermined value;
An image processing program for causing a computer to execute a process of outputting a notification when it is determined that a value extracted in a luminance value range greater than or equal to the predetermined value is larger than a value extracted in a range less than the predetermined value.

The technical features (components) described in each embodiment can be combined with each other, and new technical features can be formed by combining them.
The embodiments disclosed herein are illustrative in all respects and should not be considered as restrictive. The scope of the present invention is defined not by the above-described meaning but by the scope of the claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of the claims.

DESCRIPTION OF SYMBOLS 10 Image processing apparatus 11 Illuminating device 12 Imaging optical system 13 Imaging element 15 Imaging window 16 Illuminating window 14 Communication module 17 Cleaning device 22 Statistics unit 23 Judging unit 24 Extracting unit 25 Average unit 26 Reading unit 27 Output unit 28 Identification unit 29 Creation unit 30 server 33 camera 41 communication line 40 reception section

Claims (7)

A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
A determination unit that determines whether the luminance value extracted by the extraction unit is greater than a predetermined value;
An image processing apparatus comprising: an output unit that outputs a notification when the determination unit determines that the luminance value extracted by the extraction unit is greater than the predetermined value. An average unit for obtaining an average value of luminance values of a plurality of pixels acquired from the camera;
A determination unit for determining whether the average value is larger than a predetermined value;
An image processing apparatus comprising: an output unit that outputs a notification when the determination unit determines that the average value is larger than the predetermined value. A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
An average unit for obtaining an average value of the acquired luminance values;
A first determination unit that determines whether the luminance value extracted by the extraction unit is greater than a first predetermined value;
A second determination unit for determining whether the average value is greater than a second predetermined value;
The luminance value extracted by the extraction unit by the first determination unit is determined to be larger than the first predetermined value, and the average value is larger than the second predetermined value by the second determination unit. An image processing apparatus comprising: an output unit that outputs a notification when it is determined. The apparatus according to claim 1, further comprising: a statistical unit that obtains a statistical value of luminance values of a plurality of pixels acquired a predetermined number of times from the camera; and a setting unit that sets the predetermined value based on the statistical value. Item 4. The image processing device according to any one of Items 3 to 4. In an image processing system including an image processing apparatus and a server connected to the image processing apparatus via a communication line,
The image processing apparatus includes:
A creation unit for creating a frequency distribution of luminance values of a plurality of pixels acquired from a camera;
An extraction unit that extracts a luminance value having a maximum frequency based on the frequency distribution;
A determination unit that determines whether or not the luminance value extracted by the extraction unit is greater than a predetermined value, and a notification when the determination unit determines that the luminance value extracted by the extraction unit is greater than the predetermined value And an output unit for outputting
The server
An image processing system comprising: a reception unit that receives the notification via the communication line. Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Determining whether the extracted luminance value is greater than a predetermined value;
An image processing method comprising: outputting a notification when it is determined that the extracted luminance value is greater than the predetermined value. Create a frequency distribution of the brightness values of multiple pixels acquired from the camera,
Based on the frequency distribution, extract the luminance value with the maximum frequency,
Determining whether the extracted luminance value is greater than a predetermined value;
An image processing program for causing a computer to execute a process of outputting a notification when it is determined that the extracted luminance value is larger than the predetermined value.