Camera device and image processing method

Abstract

To adaptively adjust determination processing of a color of a subject on a captured image in accordance with setting of an image quality parameter and improving extraction accuracy of attribute information related to the color of the subject on the captured image.SOLUTION: A camera device includes: an imaging unit for capturing an image of an imaging area in which a subject is present; a memory for storing a camera parameter related to imaging; a detection unit for detecting the subject from the captured image; a first determination unit for primarily determining a color of a target portion of the detected subject; a second determination unit for adjusting a determination result of a predetermined color corresponding to the target portion determined on the basis of the camera parameter related to the imaging when the color of the target portion determined is the predetermined color having a plurality of gradations, and a communication unit for transmitting an adjustment result of the predetermined color corresponding to the determined target portion and information on the target portion in association with each other to an external device.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present disclosure relates to a camera device and an image processing method. [Background technology]

[0002] Patent Document 1 discloses a camera-equipped intercom device having a door slave unit equipped with a camera for capturing images of visitors and a function for calling residents and a room master unit equipped with a monitor for displaying the captured image of the camera and a function for responding to calls from the door slave unit. This camera-equipped intercom device creates a luminance histogram from the luminance information of the captured image of the camera, compares the luminance histogram with pre-stored reference luminance data to determine whether the captured image is in a backlight state or a low illuminance state, and generates an image by performing backlight correction and low illuminance correction. In particular, the camera-equipped intercom device performs backlight correction by performing luminance compression, which corrects the luminance of low-luminance pixels by increasing the luminance of pixels with low luminance by a predetermined amount to bring them closer to intermediate luminance, while reducing the luminance of pixels with high luminance by a predetermined amount to bring them closer to intermediate luminance, to change the luminance histogram, and performs contrast emphasis, which corrects the dark areas to be darker and the bright areas to be brighter based on a predetermined intermediate illuminance, to generate a correction tone curve. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-98726 A Summary of the Invention [Problem to be solved by the invention]

[0004] The position of a subject (e.g., a visitor) captured by the camera-equipped intercom device of Patent Document 1 tends to be a fixed position in the captured image, but the following problem occurs when trying to apply the technology of Patent Document 1 to backlight correction performed by a camera device (e.g., a surveillance camera) installed in a city or the like. Specifically, a wide variety of subjects (e.g., men, women, children, elderly people, vehicles) are captured within the angle of view of a camera device installed in a city or the like. For this reason, by uniformly changing the brightness histogram or enhancing the contrast for the entire captured image as in Patent Document 1, subjects that do not actually need to have their brightness changed or their contrast enhanced are also affected, which may cause a deterioration in visibility of the captured image.

[0005] Also, HDR (High Dynamic Range) function is known as a technology for improving the visibility of both bright areas (e.g., outdoor background objects) and dark areas (e.g., subjects such as people indoors) of an image displayed on a monitor even in places with a large difference in brightness. Camera devices that can use this HDR function have appeared in recent years, but in camera devices that can use the HDR function, the colors in the captured image may differ from the actual colors depending on the settings of the image quality parameters of the camera device. For example, when two captured images (i.e., an image captured of a bright subject and an image captured of a dark subject) captured in a backlit state are synthesized using the HDR function, an originally white subject may be processed to a light gray color, or an originally black subject may appear white.

[0006] The present disclosure has been devised in consideration of the above-mentioned conventional circumstances, and aims to provide a camera device and an image processing method that adaptively adjusts the process of determining the color of a subject in a captured image in accordance with the settings of image quality parameters, thereby improving the accuracy of extracting attribute information related to the color of the subject in the captured image. [Means for solving the problem]

[0007] The present disclosure includes an imaging unit that captures an imaging area in which a subject exists, a memory that stores camera parameters related to the imaging, a detection unit that detects the subject from an image captured by the imaging unit, a first determination unit that primarily determines a color of a site of interest of the subject detected by the detection unit, a second determination unit that adjusts a determination result of the predetermined color corresponding to the site of interest determined by the first determination unit based on the camera parameters related to the imaging when the color of the site of interest determined by the first determination unit is a predetermined color having a plurality of gradations, and a communication unit that associates the adjustment result of the predetermined color corresponding to the site of interest determined by the second determination unit with information on the site of interest and transmits them to an external device. The memory stores a color judgment table that specifies a classification ratio into classified colors corresponding to each of the plurality of gradations of the predetermined color primarily determined by the first judgment unit according to parameters related to the imaging, and the second judgment unit selects the color judgment table corresponding to the camera parameters related to the imaging, and adjusts a judgment result of the predetermined color corresponding to the site of interest determined by the first judgment unit based on any one of the selected color judgment tables, and the camera parameters related to the imaging include on or off of HDR (High Dynamic Range) and an exposure ratio for the imaging. , a camera device is provided.

[0008] The present disclosure also provides an image processing method executed by a camera device communicably connected to an external device, the method including the steps of: capturing an image of an imaging area in which a subject is present; detecting the subject from the captured image; determining a color of a portion of interest of the detected subject; and, if the determined color of the portion of interest is a predetermined color having a plurality of gradations, The memory of the camera device stores The method includes: adjusting a result of the predetermined color corresponding to the determined site of interest based on a camera parameter related to imaging; and transmitting the result of the adjustment of the predetermined color corresponding to the determined site of interest and information on the site of interest to the external device in association with each other. The memory stores a color judgment table that specifies a classification ratio of the primary determined predetermined color to a classified color corresponding to each of the multiple gradations that the primary determined predetermined color has according to parameters related to the imaging, and the adjustment of the judgment result includes selecting the color judgment table that corresponds to the camera parameters related to the imaging, and adjusting the judgment result of the predetermined color that corresponds to the determined site of interest based on any one of the selected color judgment tables, and the camera parameters related to the imaging include on or off of HDR (High Dynamic Range) and an exposure ratio for the imaging. Abstract: An image processing method is provided.

[0009] These comprehensive or specific aspects may be realized as a system, an apparatus, a method, an integrated circuit, a computer program, or a recording medium, or may be realized as any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium. Effect of the Invention

[0010] According to the present disclosure, the process of determining the color of a subject in a captured image can be adaptively adjusted according to the settings of image quality parameters, thereby improving the accuracy of extracting attribute information related to the color of the subject in the captured image.

[0011] Further advantages and benefits of certain aspects of the present disclosure will become apparent from the specification and drawings, in which such advantages and / or benefits are provided by some of the embodiments and features described in the specification and drawings, respectively, but not necessarily all of which are provided to obtain one or more identical features. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram showing an example of a system configuration of a surveillance camera system according to a first embodiment. [Diagram 2] FIG. 1 is a block diagram showing an example of a hardware configuration of a camera device according to a first embodiment; [Diagram 3] FIG. 1 is a diagram showing an example of an outline of the operation of a processor and an AI processor of a camera device according to a first embodiment; [Figure 4] A diagram showing an example of the HDR function on / off and the process of distinguishing white, gray, and black according to the exposure ratio [Figure 5A] FIG. 11 is a diagram showing an example of a color determination table when the HDR function is off. [Figure 5B] FIG. 13 is a diagram showing an example of a color judgment table when the HDR function is on and the exposure ratio is small. [Figure 5C] FIG. 13 is a diagram showing an example of a color judgment table when the HDR function is on and the exposure ratio is medium. [Figure 5D] FIG. 13 is a diagram showing an example of a color judgment table when the HDR function is on and the exposure ratio is large. [Figure 6] A flowchart showing an example of an operation procedure of the camera device according to the first embodiment. [Figure 7A] A diagram showing an example of the time change in image sensor output when the HDR function is off or the HDR function is not provided to begin with. [Figure 7B]A diagram showing an example of how the image sensor output changes over time when the HDR function is on. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] (Background to the First Embodiment) First, the problem when the HDR function of a conventional camera device is on will be described with reference to Fig. 7A and Fig. 7B. Fig. 7A is a diagram showing an example of a change in image sensor output over time when the HDR function is off or the HDR function is not originally provided. Fig. 7B is a diagram showing an example of a change in image sensor output over time when the HDR function is on. In Fig. 7A and Fig. 7B, the horizontal axis indicates time, and the vertical axis indicates the output of the image sensor of the camera device.

[0014] For example, assume that the frame rate of this camera device is 60 fps. In other words, when the HDR function of the camera device is off or the camera device does not originally have an HDR function, the camera device releases the shutter every 1 / 60 seconds to generate one captured image of the subject (see FIG. 7A).

[0015] On the other hand, when a camera device has an HDR function and the function is on, it releases the shutter every 1 / 120 seconds to capture (image) both bright and dark subjects, generating two captured images in the time it takes a camera device without an HDR function to generate one captured image (see FIG. 7B).

[0016] In Fig. 7B, the camera device releases the shutter for the first time to capture an image of a bright subject at an optimal shutter speed. The shutter speed is, for example, within a range of 1 / 120 seconds to 1 / 4000 seconds depending on the brightness of the subject. In this case, in the image captured corresponding to the first shutter speed, bright subjects are clearly displayed, but dark subjects are crushed and blacked out.

[0017] Next, the camera device releases the shutter a second time to capture an image of a dark subject at an optimal shutter speed. Because the exposure time for capturing an image of a dark subject is longer than the exposure time for capturing an image of a bright subject (see FIG. 7B), the bright subject will appear washed out (saturated), but the dark subject will be displayed clearly. The camera device can generate an image in which both the bright subject and the dark subject appear clear by combining one captured image of a bright subject and one captured image of a dark subject.

[0018] Camera devices equipped with such HDR capabilities have had the following problems:

[0019] Specifically, depending on the image quality parameters (for example, exposure ratio, which is the ratio between the exposure time for capturing a bright subject and the exposure time for capturing a dark subject) set in the camera device, depending on the imaging environment, the colors in the captured image obtained by synthesis using the HDR function may differ from the actual colors. For this reason, for example, when a subject to be monitored appears in an image captured, it may not be possible to accurately identify the colors of the subject's intended target parts (for example, clothing, hair, and vehicle body), which may reduce the efficiency of surveillance work.

[0020] For example, in a backlit shooting environment, the white parts of the subject may turn gray in the image captured by the HDR function. This is because the bright parts of a dark subject in one captured image (corresponding to the second shutter release) appear white, so when one captured image of a bright subject (corresponding to the first shutter release) and one captured image of a dark subject (corresponding to the second shutter release) are combined, the brightness of the white clothing worn by the dark subject is relatively reduced, and the subject is corrected to look as if they are wearing gray clothing.

[0021] Note that cases where the color of a captured image changes depending on the imaging environment do not necessarily occur only between white and gray. For example, when the imaging environment is low-illumination (e.g., at night), the brightness of the captured image drops overall, so that white parts may turn gray, gray parts may turn black, and orange parts may turn brown. Also, when the imaging environment is illuminated with red lighting such as halogen light, the captured image may be overall reddish, so that white parts may turn orange, and orange parts may turn red. Such color changes are not limited to the above-mentioned cases, and are a phenomenon that can occur between adjacent colors in the HSV (Hue Saturation Value) color space, for example.

[0022] Therefore, in the following embodiment 1, an example of a camera device and an image processing method are described that adaptively adjusts the process of determining the color of a subject in a captured image in accordance with the settings of image quality parameters, thereby improving the accuracy of extracting attribute information related to the color of the subject in the captured image.

[0023] Hereinafter, with reference to the drawings as appropriate, an embodiment specifically disclosing a camera device and an image processing method according to the present disclosure will be described in detail. However, more detailed explanation than necessary may be omitted. For example, detailed explanation of already well-known matters or duplicate explanation of substantially the same configuration may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to facilitate understanding by those skilled in the art. Note that the attached drawings and the following explanation are provided to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

[0024] (Embodiment 1) Fig. 1 is a diagram showing an example of a system configuration of a surveillance camera system 100 according to embodiment 1. As shown in Fig. 1, the surveillance camera system 100 includes a camera device 1 and a server 50 to which a monitor MN1 and an operation unit MH1 are connected. The camera device 1 and the server 50 are connected via a network NW1 so as to be able to transmit and receive data to and from each other.

[0025] The network NW1 is a wireless network conforming to, for example, a wireless LAN (Local Area Network) such as Wi-Fi (registered trademark), a cellular mobile communication system such as 4G or 5G, or either Bluetooth (registered trademark) or WiGig (Wireless Gigabit), but it does not have to be limited to these. Note that the network NW1 may also be a wired network such as a USB (Universal Serial Bus) cable or a wired LAN. In the following description, the image captured by the camera device 1 (hereinafter referred to as the "captured image") includes not only the data of the captured image but also the camera ID (Identification) of the camera device 1 that captured the captured image and the information of the capture date and time.

[0026] The camera device 1 is, for example, a surveillance camera installed in an imaging area AR1 such as in the street, and identifies (determines) the color of the attention part of the subject (for example, a person or a vehicle included in the angle of view from the installation position of the camera device 1) (for example, the upper body clothing, the lower body clothing, the bag of the person, the vehicle body of the vehicle). Note that the imaging area AR1, which is the installation position of the camera device 1, is not limited to the street and may also be inside a building such as a store or a building. The camera device 1 captures a captured image of the subject (see above) within the angle of view. In addition, the camera device 1 is equipped with the ability to implement artificial intelligence (AI: Artificial Intelligence), and using the installed artificial intelligence, it detects the subject (see above) from the captured captured image, or for each detected subject, it primarily determines the color of the attention part of the subject (see above) on the cut-out image (see Figure 3) obtained by cutting out the subject (details will be described later).

[0027] In addition, the camera device 1 stores a trained model obtained in advance by a learning process in accordance with the various processes by the AI ​​described above in the trained model memory 152 (see FIG. 2). The camera device 1 performs the above-mentioned detection of the subject and primary determination of the color of the attention area by executing using this trained model. This trained model is generated by a learning process by, for example, another device (not shown) or the server 50, and is stored in the camera device 1.

[0028] Furthermore, when the color determination result of the part of interest of the subject (see above) primarily determined by AI is a predetermined color (e.g., gray) having multiple gradations (see FIG. 4), the camera device 1 adjusts the primarily determined color determination result of the part of interest of the subject (see above) based on the image quality parameters (see FIG. 3) set in the camera device 1. Details of this adjustment process will be described later with reference to FIGS. 3 and 4. The camera device 1 transmits to the server 50 the primarily determined color determination result of the part of interest of the subject or the adjustment result thereof and information on the part of interest (see below) in association with at least the information.

[0029] The server 50, which is an example of an external device, is an information processing device such as a personal computer, a smartphone, a tablet terminal, or a server computer machine with high specifications. The server 50 communicates data with the camera device 1 via the network NW1.

[0030] The server 50 includes a communication interface circuit 51, a processor 52, a memory 53, and a storage 54. In the attached drawings, the interface is abbreviated to "IF" for convenience.

[0031] The communication interface circuit 51 communicates data with the camera device 1 via the network NW1 described above. The communication interface circuit 51 receives, for example, at least the primary determination result of the color of the target part of the subject or the adjustment result thereof and information on the target part (see below) transmitted from the camera device 1, and outputs them to the processor 52. Furthermore, when data of a cut-out image CT1 (see FIG. 3) in which a part of the subject is cut out is also transmitted from the camera device 1, the communication interface circuit 51 may receive data of the cut-out image CT1 (see FIG. 3). Hereinafter, the primary determination result of the color of the target part of the subject or the adjustment result thereof and information on the target part (see below), or the primary determination result of the color of the target part of the subject or the adjustment result thereof and information on the target part (see below) and the data of the cut-out image CT1 (see FIG. 3) may be collectively referred to as received data. The data of the cut-out image CT1 (see Figure 3) may be thumbnail image data that is highly compressed to the extent that the contents of the cut-out subject can be roughly identified, or it may be low-compression (including uncompressed) image data that allows the contents of the subject to be clearly identified.

[0032] The processor 52 is configured using, for example, a CPU (Central Processing Unit), a DSP (Digital Signal Processor), a GPU (Graphical Processing Unit), or an FPGA (Field Programmable Gate Array). The processor 52 functions as a controller that manages the overall operation of the server 50, and performs control processing for managing the operation of each part of the server 50, input / output processing of data between each part of the server 50, calculation processing of data, and storage processing of data. The processor 52 operates according to the programs and data stored in the memory 53. The processor 52 uses the memory 53 during operation, and temporarily stores data or information generated or acquired by the processor 52 in the memory 53.

[0033] When the processor 52 receives the reception data (see above) received by the communication interface circuit 51, it extracts from the reception data the primary determination result of the color of the part of interest of the subject or the adjustment result thereof and information on the part of interest (see later), and accumulates and saves them in the storage 54. Furthermore, when the processor 52 further extracts data of a cut-out image CT1 (see FIG. 3) from the reception data, it accumulates and saves in the storage 54 the primary determination result of the color of the part of interest of the subject or the adjustment result thereof and the information on the part of interest (see later) and the data of the cut-out image CT1 (see FIG. 3) in association with each other.

[0034] The processor 52, in cooperation with the memory 53, can also execute an application for searching and extracting data of cut-out images (e.g., image data of thumbnails that satisfy the search conditions) that satisfy the search conditions input by a user (e.g., an administrator of the surveillance camera system 100) through an operation using the operation unit MH1. The processor 52 searches the storage 54 for data of cut-out images (e.g., image data of thumbnails that satisfy the search conditions) that satisfy the search conditions (e.g., a person whose noticeable part such as clothing is white) input by a user (e.g., an administrator of the surveillance camera system 100) through an operation using the operation unit MH1, and displays the search results on the monitor MN1.

[0035] The memory 53 is configured using, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory), and temporarily stores programs necessary for executing the operation of the server 50, and further, data or information generated during operation. The RAM is, for example, a work memory used during operation of the server 50. The ROM, for example, stores and holds in advance programs for controlling the server 50.

[0036] The storage 54 is configured using, for example, a flash memory, a hard disk drive (HDD), or a solid state drive (SSD). The storage 54 accumulates and stores, for example, received data (see above) sent from the camera device 1.

[0037] The monitor MN1 is a display device configured using, for example, an LCD (Liquid Crystal Display) or an organic EL (Electroluminescence). The monitor MN1 displays, for example, the results of a search process executed by the server 50. The monitor MN1 may be included in the server 50.

[0038] The operation unit MH1 is an input device such as a mouse, keyboard, touch pad, or touch panel that accepts input operations from a user (for example, an administrator of the surveillance camera system 100). The operation unit MH1 sends a signal corresponding to the input operation of the user (for example, an administrator of the surveillance camera system 100) to the server 50. The operation unit MH1 may be included in the server 50.

[0039] Fig. 2 is a block diagram showing an example of a hardware configuration of the camera device 1 according to the embodiment 1. In addition to the camera device 1, Fig. 2 also shows an external storage medium M1 that is inserted into and removed from the camera device 1. The external storage medium M1 is, for example, a storage medium such as an SD card.

[0040] As shown in FIG. 2, the camera device 1 includes a lens 11, an image sensor 12, a memory 13, a processor 14, an AI processor 15, a communication interface circuit 16, an IR illumination unit 17, and an external storage medium interface 18.

[0041] Lens 11, which is an example of an imaging unit, includes, for example, a focus lens and a zoom lens, and receives incident light ICL1, which is light reflected by a subject, to form an optical image of the subject on the light receiving surface (imaging surface) of image sensor 12. This incident light ICL1 includes not only light reflected by a subject included in the angle of view during the day or night in imaging area AR1, such as a city where camera device 1 is installed, but also sunlight during the day. Lens 11 can have various focal lengths or imaging ranges depending on the installation location of camera device 1 or the purpose of imaging.

[0042] The image sensor 12 as an example of the imaging unit is an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). The image sensor 12 performs photoelectric conversion to convert the light received on the light receiving surface (imaging surface) into an electrical signal. Thereby, the image sensor 12 can acquire RGB (Red Green Blue) signals corresponding to the light from the subject for each pixel of a predetermined size that constitutes the light receiving surface (imaging surface). The image sensor 12 outputs an electrical signal (analog signal) corresponding to the light received on the light receiving surface to the processor 14. This analog signal is converted into data of a digital-form imaging image by the processor 14 as an example of the imaging unit. Thereby, the data of the imaging image is generated by the processor 14.

[0043] The memory 13 is configured using, for example, a RAM and a ROM, and temporarily holds programs necessary for the execution of the operation of the camera device 1, and further data or information generated during the operation. The RAM is, for example, a work memory used during the operation of the camera device 1. The ROM stores and holds in advance a program for controlling the camera device 1. In other words, the processor 14 can execute various processes that define the image processing method according to the present disclosure on the camera device 1 that is a computer by executing the program stored in the ROM.

[0044] Also, the memory 13 stores camera parameters (for example, image quality parameters) related to the imaging of the imaging unit (for example, the image sensor 12). Here, the image quality parameters are, for example, a parameter indicating whether the HDR function is on or off, and an exposure ratio (that is, when the HDR function is on, the ratio of the exposure time adjusted for imaging a bright subject at the first shutter speed to the exposure time adjusted for imaging a dark subject at the second shutter speed).

[0045] The processor 14 is configured using, for example, a CPU, a DSP, a GPU, or an FPGA. The processor 14 functions as a controller that manages the overall operation of the camera device 1, and performs control processing for managing the operation of each part of the camera device 1, input / output processing of data between each part of the camera device 1, calculation processing of data, and storage processing of data. The processor 14 operates according to the programs and data stored in the memory 13. The processor 14 uses the memory 13 during operation, and temporarily stores in the memory 13 data or information that the processor 14 generates or acquires.

[0046] The processor 14, which is an example of an imaging unit, may generate digital captured image data by performing predetermined signal processing on the electrical signal output from the imaging element 12. The processor 14 sends the generated captured image data to the AI ​​processor 15.

[0047] Processor 14 also has a timer TM1, and can ascertain the current time based on the output of timer TM1, and outputs a control signal to instruct irradiation of IR light to IR illumination unit 17 during the night (in other words, from around sunset to around dawn). Timer TM1 has a circuit that measures the elapsed time from a predetermined reference time (i.e., the current time), and sends the measurement output (count value) to processor 14. This allows processor 14 to identify the current time.

[0048] Furthermore, when the color of the portion of interest (see above) of the subject according to a color judgment AI model (see FIG. 3) described later is a predetermined color having multiple gradations (for example, a gray color having six gradations), processor 14 as an example of a second judgment unit adjusts the judgment result (i.e., the score) of the gray color of the portion of interest based on the image quality parameters stored in memory 13. Details of this adjustment process will be described later with reference to FIGS. 3 and 4.

[0049] The AI ​​processor 15 includes an AI calculation processing unit 151 and a learning model memory 152. That is, the AI ​​processor 15 selects and uses a learned model stored in the learning model memory 152, and causes the AI ​​calculation processing unit 151 to execute a specific process corresponding to the learned model.

[0050] The AI ​​calculation processing unit 151 is configured using, for example, a CPU, DSP, GPU or FPGA, and selects and uses one of the trained models stored in the training model memory 152, and executes processing specific to that trained model.

[0051] For example, the AI ​​calculation processing unit 151, which is an example of a detection unit, reads out an AI trained model for detecting subjects such as people or vehicles (i.e., the object detection AI model shown in FIG. 3) from the trained model memory 152, and executes it to detect and extract the subject to be monitored from the data of the captured image captured by the image sensor 12.

[0052] For example, the AI ​​arithmetic processing unit 151 as an example of a first judgment unit reads out an AI trained model (i.e., a color judgment AI model shown in FIG. 3) for primarily judging the color of a noticeable part (e.g., clothing, hair, vehicle body) of a subject such as a person or a vehicle from the trained model memory 152, executes it, and primarily judges the color of the noticeable part (e.g., clothing, hair, vehicle body) of the subject detected from the data of the cropped image after resizing (see FIG. 3) by the AI ​​arithmetic processing unit 151. Each time a subject is detected from the data of the captured image captured by the imaging element 12, the AI ​​arithmetic processing unit 151 primarily judges the color of the noticeable part (e.g., clothing, hair, vehicle body) of the subject. Details of this primary judgment process will be described later with reference to FIG. 3.

[0053] The learning model memory 152 is configured with a memory such as a RAM, a ROM, a flash memory, etc. The learning model memory 152 stores a learned model (for example, the above-mentioned object detection AI model and color judgment AI model) that has been created in advance by a learning process.

[0054] The communication interface circuit 16, which is an example of a communication unit, performs data communication (transmission and reception) with the server 50 connected via the network NW1. The communication interface circuit 16 transmits, for example, received data generated by the processor 14 (that is, the primary determination result of the color of the target part of the subject or the adjustment result thereof and information on the target part (see below), or the primary determination result of the color of the target part of the subject or the adjustment result thereof and information on the target part (see below) and data on the cut-out image CT1 (see FIG. 3)) to the server 50. In addition, when the communication interface circuit 16 receives camera parameters (for example, parameters indicating whether the above-mentioned HDR function is on or off, and exposure ratio) from the server 50, the communication interface circuit 16 stores the camera parameters in the memory 13.

[0055] The IR illumination unit 17 starts irradiating the imaging area with IR light RD1 having a near-infrared wavelength band based on a control signal (e.g., an instruction to start irradiating IR light) from the processor 14. The IR illumination unit 17 ends the irradiation of the IR light RD1 to the imaging area based on a control signal (e.g., an instruction to end irradiation of IR light RD1) from the processor 14. Furthermore, the IR illumination unit 17 irradiates the currently irradiated IR light RD1 by increasing or decreasing the intensity based on a control signal (e.g., an instruction to adjust the intensity of IR light RD1) from the processor 14.

[0056] An external storage medium M1 such as an SD card is inserted into and removed from the external storage medium interface 18.

[0057] Next, an example of an outline of the operation of the processor 14 and the AI ​​processor 15 of the camera device 1 according to the embodiment 1 will be described with reference to Fig. 3 and Fig. 4. Fig. 3 is a diagram showing an example of an outline of the operation of the processor 14 and the AI ​​processor 15 of the camera device 1 according to the embodiment 1. Fig. 4 is a diagram showing an example of an outline of the process of distinguishing white, gray, and black according to on / off of HDR and exposure ratio.

[0058] 3, the processor 14 generates a captured image IMG1 of an imaging area AR1 captured by the imaging element 12. Here, the size of the captured image IMG1 is, for example, 1920 [dpi]×1080 [dpi], but is not limited to this size. The processor 14 reduces (resizes) the size of the captured image IMG1 and sends it to the AI ​​processor 15 (St1) so that it can be input to an object detection AI model executed by the AI ​​processor 15 described later. For example, the processor 14 resizes the captured image IMG1 to reduce its size to 1024 [dpi]×1024 [dpi].

[0059] The AI ​​processor 15 reads out an object detection AI model from the learning model memory 152, for example, executes it, detects a subject (e.g., a person) from the captured image resized in step St1, and sends information about the subject (e.g., the subject's position, gender, age, etc.) to the processor 14 (St2). For example, the size of the subject detected by the object detection AI model is 200 [dpi] × 400 [dpi].

[0060] The processor 14, as an example of a cutout processing unit, generates a cutout image CT1 by cutting out an outer portion including the contour of the subject from the captured image IMG1 before the resizing process of step St1, using the captured image IMG1 before the resizing process of step St1 and information about the subject from the AI ​​processor 15 in step St2 (see above) (St3). Furthermore, the processor 14 resizes the size of this cutout image CT1 so that it can be input to a color judgment AI model executed by the AI ​​processor 15 described later, and sends it to the AI ​​processor 15 (St3). For example, the processor 14 resizes the size of the cutout image CT1 to 224 [dpi] x 224 [dpi].

[0061] The AI ​​processor 15 reads out and executes a color judgment AI model from the learning model memory 152, for example, and judges the color of the subject's attention part (for example, clothing of the upper body and clothing of the lower body) from the cropped image of the subject resized in step St3, and sends the judgment result to the processor 14 (St4). The AI ​​processor 15 uses the classification label LBL1 consisting of a predetermined number of color items as the output of step St4, for example, to calculate the probability that the color of the subject's attention part (see above) is closest to each of the colors constituting the classification label LBL1 for each color of the classification label LBL1. This classification label LBL1 is determined in advance during the learning process of the color judgment AI model. The classification label LBL1 is composed of, for example, black, brown, white, gray having six gradations, green, red, blue, yellow, orange, purple, and pink.

[0062] In step St4, when the color of the part of interest of the subject (see above) by the color judgment AI model is a predetermined color having multiple gradations (for example, a gray color having six gradations), the processor 14 adjusts the judgment result of the gray color of the part of interest (i.e., the score) based on the camera parameters CP1 (for example, the image quality parameters described above) stored in the memory 13 (St5). That is, as described above, in a backlight state, the white part of the subject (for example, white clothes worn by a person) may change to gray on the captured image captured when the HDR function is on. Therefore, in order to reduce the difference between the actual color and the color classification result due to such a color change on the captured image, the score corresponding to the part judged as gray by the color judgment AI model is adjusted according to the camera parameters CP1. The processor 14 organizes and summarizes the score adjustment results performed in step St5 as the final classification score SCR1.

[0063] Next, an example of the outline of the process of the processor 14 in step St5 described above will be described with reference to Fig. 4 and Figs. 5A, 5B, 5C, and 5D. Fig. 5A is a diagram showing an example of a color judgment table TBL1 when the HDR function is off. Fig. 5B is a diagram showing an example of a color judgment table TBL2 when the HDR function is on and the exposure ratio is small. Fig. 5C is a diagram showing an example of a color judgment table TBL3 when the HDR function is on and the exposure ratio is medium. Fig. 5D is a diagram showing an example of a color judgment table TBL4 when the HDR function is on and the exposure ratio is large.

[0064] In the example of Figure 4, a case is illustrated in which the color judgment AI model judges that the color of a focal part (e.g., clothing of the upper body) of a subject (e.g., a male person) is a gray color having six gradations. For example, the six gradations are Gray1, Gray2, Gray3, Gray4, Gray5, and Gray6, with Gray1 being closest to white and Gray6 being closest to black.

[0065] In Figure 4, it is assumed that the color judgment AI model has obtained a score GRD1 for the color items White, Gray1, Gray2, Gray3, Gray4, Gray5, Gray6, and Black.

[0066] When the camera parameter CP1 stored in the memory 13 indicates that the HDR function is off, the processor 14 reads out from the memory 13 a color judgment table TBL1 corresponding to the HDR function being off. The color judgment table TBL1 is used when the HDR function of the camera device 1 is off, and is a table that specifies a weighting factor (specific gravity) for assigning each color item (i.e., classification label) calculated by the color judgment AI model to one of black, gray, and white. The processor 14 calculates the multiplication result of the weighting factor (specific gravity) for each classification label specified in the color judgment table TBL1 and the score of each classification label for each of black, gray, and white, derives the calculated score RST1 as an adjustment result RST5 of the final classification score SCR1, and sends it to the server 50 as a judgment result of the color of the target area.

[0067] For example, the processor 14 calculates the following for black: (score of classification label "Black" x weighting factor "1.0") + (score of classification label "Gray6" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for gray: (score of classification label "Gray6" x weighting factor "0.5") + (score of classification label "Gray5" x weighting factor "1.0") + (score of classification label "Gray4" x weighting factor "1.0") + (score of classification label "Gray3" x weighting factor "1.0") + (score of classification label "Gray2" x weighting factor "1.0") + (score of classification label "Gray1" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for white: (score of classification label "Gray1" x weighting factor "0.5") + (score of classification label "White" x weighting factor "1.0"). The calculation results of these three colors, black, gray, and white, are derived by the processor 14 as an adjustment result RST5 of the final classification score SCR1.

[0068] Furthermore, when the camera parameters CP1 stored in the memory 13 indicate that the HDR function is on and the exposure ratio (see above) is less than the threshold value Th1, the processor 14 reads out from the memory 13 a color judgment table TBL2 corresponding to the HDR function being on and the exposure ratio being small. The color judgment table TBL2 is used when the HDR function of the camera device 1 is on and the exposure ratio is less than the threshold value Th1, and is a table that specifies a weighting factor (weight) for assigning each color item (i.e., classification label) calculated by the color judgment AI model to one of black, gray, and white. The processor 14 calculates the multiplication result of the weighting factor (weight) for each classification label specified in the color judgment table TBL2 and the score of each classification label for each of black, gray, and white, derives the calculated score RST2 as the adjustment result RST5 of the final classification score SCR1, and sends it to the server 50 as the judgment result of the color of the target area.

[0069] For example, the processor 14 calculates the following for black: (score of classification label "Black" x weighting factor "1.0") + (score of classification label "Gray6" x weighting factor "0.7") + (score of classification label "Gray5" x weighting factor "0.3"). Similarly, the processor 14 calculates the following for gray: (score of classification label "Gray5" x weighting factor "0.7") + (score of classification label "Gray4" x weighting factor "1.0") + (score of classification label "Gray3" x weighting factor "1.0") + (score of classification label "Gray2" x weighting factor "1.0") + (score of classification label "Gray1" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for white: (score of classification label "Gray1" x weighting factor "0.5") + (score of classification label "White" x weighting factor "1.0"). The calculation results of these three colors, black, gray, and white, are derived by the processor 14 as an adjustment result RST5 of the final classification score SCR1.

[0070] Furthermore, when the camera parameters CP1 stored in the memory 13 indicate that the HDR function is on and the exposure ratio (see above) is between the threshold value Th1 and the threshold value Th2, the processor 14 reads out from the memory 13 a color judgment table TBL3 corresponding to the HDR function being on and the exposure ratio being medium. The color judgment table TBL3 is used when the HDR function of the camera device 1 is on and the exposure ratio is between the threshold value Th1 and the threshold value Th2, and is a table that specifies a weighting factor (weight) for assigning each color item (i.e., classification label) calculated by the color judgment AI model to one of black, gray, and white. The processor 14 calculates the multiplication result of the weighting factor (weight) for each classification label specified in the color judgment table TBL3 and the score of each classification label for each of black, gray, and white, derives the calculated score RST3 as the final adjustment result RST5 of the classification score SCR1, and sends it to the server 50 as the judgment result of the color of the target area.

[0071] For example, the processor 14 calculates the following for black: (score of classification label "Black" x weighting factor "1.0") + (score of classification label "Gray6" x weighting factor "1.0") + (score of classification label "Gray5" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for gray: (score of classification label "Gray5" x weighting factor "0.5") + (score of classification label "Gray4" x weighting factor "1.0") + (score of classification label "Gray3" x weighting factor "1.0") + (score of classification label "Gray2" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for white: (score of classification label "Gray2" x weighting factor "0.5") + (score of classification label "Gray1" x weighting factor "1.0") + (score of classification label "White" x weighting factor "1.0"). The calculation results of these three colors, black, gray, and white, are derived by the processor 14 as an adjustment result RST5 of the final classification score SCR1.

[0072] Furthermore, when the camera parameters CP1 stored in the memory 13 indicate that the HDR function is on and the exposure ratio (see above) is equal to or greater than the threshold Th2, the processor 14 reads out from the memory 13 a color judgment table TBL4 corresponding to the HDR function being on and the large exposure ratio. The color judgment table TBL4 is used when the HDR function of the camera device 1 is on and the exposure ratio is equal to or greater than the threshold Th2, and is a table that specifies a weighting factor (weight) for assigning each color item (i.e., classification label) calculated by the color judgment AI model to one of black, gray, and white. The processor 14 calculates the multiplication result of the weighting factor (weight) for each classification label specified in the color judgment table TBL4 and the score of each classification label for each of black, gray, and white, derives the calculated score RST4 as the final adjustment result RST5 of the classification score SCR1, and sends it to the server 50 as the judgment result of the color of the target area.

[0073] For example, the processor 14 calculates the following for black: (score of classification label "Black" x weighting factor "1.0") + (score of classification label "Gray6" x weighting factor "1.0") + (score of classification label "Gray5" x weighting factor "0.5"). Similarly, the processor 14 calculates the following for gray: (score of classification label "Gray5" x weighting factor "0.5") + (score of classification label "Gray4" x weighting factor "1.0") + (score of classification label "Gray3" x weighting factor "0.7"). Similarly, the processor 14 calculates the following for white: (score of classification label "Gray3" x weighting factor "0.3") + (score of classification label "Gray2" x weighting factor "1.0") + (score of classification label "Gray1" x weighting factor "1.0") + (score of classification label "White" x weighting factor "1.0"). The calculation results of these three colors, black, gray, and white, are derived by the processor 14 as an adjustment result RST5 of the final classification score SCR1.

[0074] Next, an example of an operation procedure of the camera device 1 according to the embodiment 1 will be described with reference to Fig. 6. Fig. 6 is a flowchart showing an example of an operation procedure of the camera device 1 according to the embodiment 1. Each process shown in Fig. 6 is executed by the processor 14 or the AI ​​processor 15 of the camera device 1.

[0075] 6, the image sensor 12 captures an image of a subject in the imaging area AR1 in which the camera device 1 is installed (St11). The processor 14 applies a predetermined signal processing to the electric signal output from the image sensor 12 to generate digital captured image data and send it to the AI ​​processor 15. At this time, the processor 14 reads and acquires the on / off state of the HDR function and the exposure ratio as an example of the camera parameter CP1 related to imaging from the memory 13 (St12). The camera parameter CP1 may be set and stored in the memory 13 when the camera device 1 is initially installed in the imaging area AR1, for example, or may be stored in the memory 13 in response to an instruction from the server 50.

[0076] The AI processor 15 resizes (see FIG. 3) the data of the captured image generated in step St11 so as to conform to the input of an object detection AI model (an example of a learned model) for detecting a subject such as a person or a vehicle (St13). Further, the AI processor 15 reads out and executes the object detection AI model from the learning model memory 152, and detects and extracts a subject to be monitored (for example, a person or a vehicle) from the data of the captured image after resizing in step St13 (St13). If no subject (for example, a person or a vehicle) is detected (St14, NO), the process of the camera device 1 returns to step St11.

[0077] On the other hand, when a subject (for example, a person or a vehicle) is detected by the AI processor 15 (St14, YES), the processor 14 uses the position (for example, coordinates) of the subject (for example, a person or a vehicle) on the resized captured image input to the object detection AI model to cut out the area of the subject (for example, a person or a vehicle) from the resized captured image to generate a cut-out image (for example, the cut-out image CT1 in FIG. 3) and send it to the AI processor 15 (St15).

[0078] The AI processor 15 resizes (see FIG. 3) the data of the cut-out image generated in step St15 so as to conform to the input of a color determination AI model (an example of a learned model) for primarily determining the color of a notable part (for example, clothing, hair, vehicle body) of a subject such as a person or a vehicle (St16). Further, the AI processor 15 reads out and executes the color determination AI model from the learning model memory 152, primarily determines the color of the notable part (for example, clothing, hair, vehicle body) of the detected subject from the data of the cut-out image after resizing in step St16, and sends the determination result to the processor 14 (St16).

[0079] The processor 14 refers to the memory 13 and determines whether or not the HDR function, which is an example of the camera parameter CP1 of the camera device 1, is on (St17). If the processor 14 determines that the HDR function is not on (i.e., off) (St17, NO), it selects and reads from the memory 13 the color determination table TBL1 corresponding to the HDR function being off (St18).

[0080] On the other hand, if the processor 14 determines that the HDR function is on (St17, YES), it selects one of the color judgment tables (specifically, one of the color judgment tables TBL1, TBL2, or TBL3) that corresponds to the HDR function being on and reads it from the memory 13 (St19).

[0081] The processor 14 calculates and adjusts (St20) the classification score of the color of the object of interest in the cropped image after resizing in step St16, using the primary color judgment result by the color judgment AI model in step St16 and the color judgment table selected in step St18 or step St19. An example of the adjustment (calculation) in step St20 has been described with reference to FIG. 5, so a description thereof will be omitted here.

[0082] The processor 14 determines whether the color classification score SCR1 (in other words, the final color judgment result) resulting from the adjustment in step St20 exceeds a predetermined threshold value (stored in the memory 13) prepared for each color item (St21). If it is determined that the color classification score SCR1 is less than the predetermined threshold value (e.g., 0.3) prepared for each color item (St21, NO), the process of the camera device 1 returns to step St11.

[0083] On the other hand, if the processor 14 determines that the color classification score SCR1 is equal to or greater than a predetermined threshold value (e.g., 0.3) prepared for each color item (St21, YES), it transmits to the server 50 the color classification score SCR1 calculated in step St20, information on the focal part of the subject (e.g., a person or vehicle) (e.g., clothing, hair, body), and the cut-out image generated in step St16 (St22).

[0084] As described above, in the surveillance camera system 100 according to the first embodiment, the camera device 1 includes an imaging unit which captures an image of an imaging area AR1 in which a subject (e.g., a person or a vehicle) is present, a memory 13 which stores camera parameters CP1 relating to imaging, a detection unit (e.g., AI processor 15) which detects the subject from an image captured by the imaging unit, a first judgment unit (e.g., AI processor 15) which primarily judges the color of a feature of the subject detected by the detection unit (e.g., clothing, hair, vehicle body), a second judgment unit (e.g., processor 14) which, when the color of the feature determined by the first judgment unit is a predetermined color having multiple gradations (e.g., a gray color having six gradations), adjusts the judgment result of the predetermined color corresponding to the feature determined by the first judgment unit based on the parameters relating to imaging, and a communication unit (e.g., communication interface circuit 16) which associates the adjustment result of the predetermined color corresponding to the feature determined by the second judgment unit with information on the feature and transmits them to an external device (e.g., server 50).

[0085] As a result, the camera device 1 can adaptively adjust the color judgment process of the subject in the captured image even in a backlit state, for example, according to the contents of the image quality parameters (i.e., camera parameters CP1) set in the camera device 1. Therefore, the camera device 1 can improve the extraction accuracy of attribute information on the color of the subject in the captured image (in other words, the color of the part of the subject that should be noted). In addition, since the conventional technology could not respond to the color change in the captured image when the environment changes during imaging (for example, when the image becomes backlit), the judgment accuracy of the color of the part of the subject that should be noted deteriorates, or a dedicated color judgment AI model needs to be constructed for each scene in which a color change occurs, which increases the development man-hours. However, according to the camera device 1 according to the first embodiment, the score can be corrected by the color judgment AI model according to the camera parameter CP1, so that color judgment that can follow the environmental change during imaging can be performed without constructing multiple color judgment AI models according to the type of environment during imaging, and deterioration of the accuracy of color judgment due to the environmental change during imaging can be suppressed.

[0086] Moreover, the first determination unit primarily determines the color of the attention portion of the subject every time the subject is detected by the detection unit. This allows the camera device 1 to continuously determine the color of the attention portion of the detected subject every time the subject (e.g., a person or a vehicle) is detected in the imaging area AR1.

[0087] The camera device 1 further includes a cutout processing unit (e.g., processor 14) that generates a cutout image (e.g., cutout image CT1) by cutting out the subject detected by the detection unit from the captured image. The communication unit transmits to an external device the adjustment result of the predetermined color corresponding to the part of interest determined by the second determination unit, information on the part of interest, and the cutout image in association with each other. This allows the external device (e.g., server 50) to collectively receive and store the reception data sent from the camera device 1 (i.e., the primary determination result of the color of the part of interest of the subject or the adjustment result, information on the part of interest, and data on the cutout image).

[0088] Furthermore, the memory 13 stores color judgment tables TBL1-TBL4 that define the classification ratio into classified colors (e.g., gray or white) corresponding to each of a plurality of gradations of the predetermined color primarily judged by the first judgment unit according to the camera parameter CP1 related to imaging. The second judgment unit selects a color judgment table corresponding to the camera parameter CP1 related to imaging, and adjusts the judgment result of the predetermined color corresponding to the attention site judged by the first judgment unit based on one of the selected color judgment tables. This allows the camera device 1 to correct the color of the original attention site of the subject even if a change occurs (e.g., white changes to gray) when two captured images are combined using the HDR function to capture a bright subject and a dark subject in a backlit state, for example, so that the color can be accurately judged as the color of the attention site of the subject.

[0089] The camera parameters CP1 related to imaging include whether an HDR (High Dynamic Range) function is on or off and an exposure ratio for imaging. This allows the camera device 1 to adaptively select a color judgment table according to whether the HDR function is on or off and the exposure ratio for imaging.

[0090] The first determination unit also outputs a score indicating the likelihood of each classification label of the color of the region of interest as a primary determination result of the color of the region of interest through AI processing using a trained model for primarily determining the color of the region of interest. This allows the camera device 1 to determine the color of the region of interest of the subject with high accuracy.

[0091] The target part is the clothes or hair of a person present in the imaging area AR1, or the body of a vehicle present in the imaging area AR1. This allows the camera device 1 to determine the color of the target part, such as the clothes or hair of a person present in the imaging area AR1, or the body of a vehicle present in the imaging area AR1, with high accuracy, in accordance with the environment at the time of imaging.

[0092] Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can conceive of various modifications, corrections, substitutions, additions, deletions, and equivalents within the scope of the claims, and it is understood that these also naturally belong to the technical scope of the present disclosure. In addition, the components in the various embodiments described above may be arbitrarily combined within the scope of the invention. [Industrial Applicability]

[0093] The present disclosure is useful as a camera device and an image processing method that adaptively adjusts the process of determining the color of a subject in a captured image in accordance with the settings of image quality parameters, thereby improving the accuracy of extracting attribute information related to the color of the subject in the captured image. [Explanation of symbols]

[0094] 1 Camera equipment 11 Lens 12 Image sensor 13,53 Memory 14, 52 processors 15 AI Processor 16, 51 Communication interface circuit 17 IR lighting section 18 External storage media interface 50 Servers 54 Storage 100 Surveillance Camera System 151 AI processing unit 152 Learning Model Memory M1 external storage medium MH1 operation unit MN1 Monitor

Claims

1. an imaging unit that captures an image of an imaging area in which a subject is present; A memory for storing camera parameters related to the imaging; a detection unit that detects the subject from an image captured by the imaging unit; a first determination unit that primarily determines a color of the attention portion of the subject detected by the detection unit; a second determination unit that adjusts a determination result of the predetermined color corresponding to the portion of interest determined by the first determination unit based on a camera parameter related to the imaging when the color of the portion of interest determined by the first determination unit is a predetermined color having a plurality of gradations; a communication unit that associates an adjustment result of the predetermined color corresponding to the portion of interest determined by the second determination unit with information of the portion of interest and transmits the result to an external device, the memory stores a color judgment table that specifies a classification ratio of the predetermined color primarily judged by the first judgment unit into a classified color corresponding to each of the plurality of gradations of the predetermined color according to a parameter related to the imaging; the second determination unit selects the color determination table corresponding to a camera parameter related to the imaging, and adjusts a determination result of the predetermined color corresponding to the site of interest determined by the first determination unit based on any one of the selected color determination tables; The camera parameters related to the imaging include on or off of HDR (High Dynamic Range) and an exposure ratio of the imaging. Camera equipment.

2. the first determination unit primarily determines a color of a noteworthy portion of the subject every time the subject is detected by the detection unit; The camera device according to claim 1 .

3. a cut-out processing unit that generates a cut-out image by cutting out the subject detected by the detection unit from the captured image, the communication unit transmits to the external device an adjustment result of the predetermined color corresponding to the portion of interest determined by the second determination unit, information on the portion of interest, and the cut-out image in association with each other. The camera device according to claim 1 .

4. The first determination unit outputs a score indicating a likelihood of each classification label of the color of the attention area as a primary determination result of the color of the attention area by AI (Artificial Intelligence) processing using a trained model for primarily determining the color of the attention area. The camera device according to claim 1 .

5. the attention part is clothing or hair of a person present in the imaging area, or a body of a vehicle present in the imaging area; The camera device according to claim 1 .

6. An image processing method executed by a camera device communicatively connected to an external device, comprising: capturing an image of an imaging area in which a subject is present; detecting the subject from a captured image; A step of primarily determining a color of a portion of interest of the detected subject; a step of adjusting a result of the determination of the predetermined color corresponding to the determined portion of interest based on a camera parameter related to imaging stored in a memory of the camera device when the determined color of the portion of interest is a predetermined color having a plurality of gradations; and transmitting the adjustment result of the predetermined color corresponding to the determined portion of interest and information about the portion of interest to the external device in association with each other, the memory stores a color determination table that defines a classification ratio of the primary determined predetermined color into a classified color corresponding to each of the plurality of gradations in accordance with a parameter related to the imaging; In the adjustment of the judgment result, the color judgment table corresponding to the camera parameters related to the imaging is selected, and the judgment result of the predetermined color corresponding to the determined part of interest is adjusted based on any one of the selected color judgment tables; The camera parameters related to the imaging include on or off of HDR (High Dynamic Range) and an exposure ratio of the imaging. Image processing methods.

Images