JP6727998B2 - Image processing apparatus, image processing method and program - Google Patents

Description

The present invention relates to an image processing device, an image processing method, and a program.

Conventionally, there is known a technique of recognizing a specific subject from an image obtained by an imaging device. Patent Document 1 discloses a technique of recognizing a preset specific subject from the entire image and displaying a frame around the recognized specific subject.

JP, 2008-78951, A

However, for example, in an image obtained by using a wide-angle lens or a fish-eye lens, there is a problem that it may not be possible to perform accurate recognition in a partial area of the image due to image distortion.

The present invention has been made in view of such problems, and an object thereof is to improve the recognition accuracy of a specific subject in an image with distortion.

Therefore, the present invention is an image processing device, and a designating unit that designates an application range in which a specific subject is recognized from an image obtained by an imaging unit, and a reference range that can designate the application range by the designating unit. Is displayed on the display means.

According to the present invention, it is possible to improve the recognition accuracy of a specific subject in an image with distortion.

It is a general view of a camera system. It is explanatory drawing of the process of a recognition part. It is a flow chart which shows image processing. It is an explanatory view of image processing. It is an explanatory view of image processing. It is a general view of a camera system according to a second embodiment. 9 is a flowchart showing image processing according to the second embodiment. It is an explanatory view of image processing.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is an overall view of a camera system 100 according to the first embodiment. The camera system 100 has a camera 110 as an imaging device, a management device 120, and a display unit 130. Here, the camera 110 and the management device 120 are both examples of an image processing device. The camera 110 includes a lens 111, an image pickup unit 112, a video processing unit 113, a development unit 114, a recognition unit 115, a communication unit 116, a CPU 117, and a storage unit 118. The lens 111 is a main lens and is a fisheye lens in this embodiment. The lens 111 may be a wide-angle lens, a lens having other characteristics, or the like as long as it can obtain an image in which the image is distorted, and is not limited to a fisheye lens.

The imaging unit 112 converts the light incident through the lens 111 into a video (image). The video processing unit 113 performs various video processing for improving the image quality of the video imaged by the imaging unit 112. The expansion unit 114 expands and corrects the image processed by the image processing unit 113 into an image optimal for recognition processing.

The recognition unit 115 performs recognition processing of a specific subject on the video developed by the video processing unit 113 using a recognition dictionary set from the outside. Here, the recognition dictionary is data related to an algorithm for recognizing the characteristics of a specific subject. Further, in the present embodiment, the specific subject is assumed to be a person. The type of the specific subject is not limited to the embodiment, and the recognition unit 115 may recognize a subject other than a person. In the present embodiment, the recognition unit 115 can perform the first recognition process, the second recognition process, and the third recognition process. The recognition process will be described later with reference to FIG. The communication unit 116 communicates with an external device via the network 150. The communication unit 116 transmits, for example, a video image processed by the video processing unit 113 to the external device in response to a command requested by the external device.

The CPU 117 controls the entire camera 110. The storage unit 118 stores various kinds of information. The video processing unit 113, the expansion unit 114, and the recognition unit 115 according to this embodiment are assumed to be independent hardware. However, as another example, the image processing unit 113, the expansion unit 114, and the recognition unit 115 have a function realized as software by the CPU 117 reading a program stored in the storage unit 118 and executing the program. It may be.

The management device 120 manages the camera 110. The management device 120 performs, for one camera 110, a real-time captured video request instruction, a recognition range designation, system management control, processing in response to a warning from each device, and the like. The display unit 130 displays an image or various information transmitted from the camera 110 via the network 150 in response to a request from the management device 120.

FIG. 2 is an explanatory diagram of the processing of the recognition unit 115. FIG. 2A is a diagram showing an example of an image (image) 200 when the camera 110 having a fisheye lens is installed on the ceiling. In the image 200, in the central area 210 including the center of the image, as in 201, the person only sees the head. On the other hand, in the outer area 220, which is an area outside the central area 210, as shown at 202, the body other than the head, such as the torso and legs of the person, can be seen, but the image also has distortion. In such a case, if the same recognition process is performed on the central area 210 and the outer area 220, the recognition cannot be performed accurately.

Therefore, the recognition unit 115 performs the first recognition process suitable for the central region 210 on the central region 210, and performs the second recognition process suitable for the outer region 220 on the outer region 220. I will. That is, the first recognition process is a recognition process suitable for recognizing a person from an image obtained by shooting only the head in a vertically downward direction, and is a process using the corresponding first recognition dictionary. The second recognition process is a recognition process suitable for recognizing a person from an image of a person including a torso and the like, which is particularly distorted in the height direction of the person. The second recognition process is a process using the second recognition dictionary.

Further, in the camera 110, the first reference range is preset as the reference range suitable for performing the first recognition process, and the first reference range is suitable as the reference range for performing the second recognition process. It is assumed that the reference range of 2 is set in advance. Specifically, it is assumed that the storage unit 118 of the camera 110 stores the relative positions of the first reference range and the second reference range with respect to the entire image. Here, the first reference range is a range in which recognition can be performed accurately by the first recognition process, and can be specified by the user as a range to be a target (recognition target) of the first recognition process. .. The second reference range is a range in which recognition can be performed accurately by the second process, and is a range that can be designated by the user as a range (recognition target) of the second recognition process.

FIG. 2B is a diagram showing an example of a video (image) 230 when the camera 110 is installed on the horizontal wall. In the image 230, a person including a body like 231 is shown, but the image is distorted. However, this distortion is a different kind of distortion from the distortion of the outer region 220 of the image 200. Therefore, the recognition unit 115 performs the third recognition process on the image 200. Here, the third recognition process is a recognition process suitable for recognizing a person from an image of a person including a torso and the like, in which the image is particularly distorted in a direction perpendicular to the height direction of the person.

Further, in the image 230, the range of the curves 241 to 242 corresponds to the wall, and the range of the curves 242 to 243 corresponds to the floor. The upper side of the curve 241 corresponds to the ceiling. Therefore, the person exists in the range between the curves 241 and 243. Therefore, it is assumed that the storage unit 118 of the camera 110 is preset with a third reference range determined based on the curves 241 to 243 as a range suitable for performing the third recognition process. Note that the first recognition process, the second recognition process, and the third recognition process are different from each other. The first reference range, the second reference range, and the third reference range are different from each other.

FIG. 3 is a flowchart showing an image processing operation by the camera 110. When the user turns on the power after mounting the camera system 100 at a desired place, the image processing by the camera 110 is started. In S100, the CPU 117 determines whether or not it is in the installation mode according to the input from the user. The CPU 117 stands by when it is not in the installation mode (NO in S100), and when it is in the installation mode (YES in S100), advances the process to S101.

In S101, the CPU 117 receives the setting regarding the network 150 set by the user in the management apparatus 120 via the communication unit 116, and sets the setting. The CPU 117 performs necessary settings for network connection conditions that match the installation environment, such as setting an IP address for identifying individual communication on the network communication. Next, in S102, the CPU 117 controls the lens 111 and the imaging unit 112 to start shooting. Then, the CPU 117 distributes the video to the management device 120 according to the network settings set in S101. When the management device 120 receives the video, the management device 120 displays the video on the display unit 130.

Next, in S103, the CPU 117 confirms whether the installation location of the camera 110 is the ceiling or the lateral wall. The user inputs the installation location to the management device 120. Then, the CPU 117 receives the installation information indicating the installation location from the management device 120. If the installation location indicated by the installation information is the ceiling (YES in S103), CPU 117 advances the process to S104. CPU117 advances a process to S115, when an installation place is a horizontal wall (NO in S103). In S104, the CPU 117 sets the installation place to the ceiling.

Next, in S105, the CPU 117 sets shooting parameters so that an optimum image corresponding to the installed shooting environment can be obtained. Here, the shooting parameter is a parameter related to image quality. For example, in a place where the environment is dark, the CPU 117 sets parameters for AE control so that the image becomes bright. Further, when it is desired to make the image vivid, the CPU 117 sets a parameter for adjusting the saturation. The CPU 117 sets shooting parameters under the control of the management device 120.

Next, in S106, the CPU 117 confirms whether or not to perform the first recognition process. Note that the management device 120 receives an instruction from the user as to whether or not to perform the first recognition process, and the CPU 117 receives the instruction from the management device 120. The CPU 117 confirms whether to perform the first recognition process based on the received instruction. That is, when the CPU 117 receives an instruction to perform the first recognition process, it determines that the first recognition process is to be performed, and when it receives an instruction to not perform the first recognition process. , It is determined that the first recognition process is not performed. When performing the first recognition process (YES in S106), CPU 117 advances the process to S107. CPU117 advances a process to S109, when the 1st recognition process is not performed (it is NO at S106).

In S107, the CPU 117 specifies the first recognition process and the second recognition process based on the installation place (ceiling) set in S104. Then, the CPU 117 specifies the first reference range corresponding to the first recognition process in the video as the recognition range of the recognition unit 115, and the specified first reference range is displayed on the display unit 130 via the management device 120. Control to display. Next, in S108, the CPU 117 sets a first application range to which the first recognition process is applied, according to the operation of the setting frame by the user. Here, the first application range is a range set according to a user operation as a range to which the first recognition process is applied. In this way, the user can set a desired range as the first applicable range within the first reference range.

Next, in S109, the CPU 117 confirms whether or not to perform the second recognition process based on the instruction received from the management device 120. When performing the second recognition process (YES in S109), CPU 117 advances the process to S110. CPU117 advances a process to S112, when the 2nd recognition process is not performed (NO in S109).

In S110, the CPU 117 specifies the second reference range corresponding to the second recognition process as the recognition range of the recognition unit 115. Then, the CPU 117 controls the display unit 130 to display the second reference range via the management device 120. Next, in S111, the CPU 117 sets a second application range to which the second recognition process is applied, in accordance with the operation of the setting frame by the user. Here, the second applicable range is a range set according to a user operation as a range in which the second recognition process is performed. In this way, the user can set a desired range as the second applicable range within the second reference range.

Next, in S112, the CPU 117 confirms whether or not the set applicable range is changed. When there is no change (YES in S112), the CPU 117 ends the image processing. When there is a change (NO in S112), CPU 117 advances the process to S113. In S113, the CPU 117 confirms whether or not to reset the applicable range. When resetting (YES in S113), CPU 117 advances the process to S106. CPU117 advances a process to S114, when not resetting (it is NO at S113). In S114, the CPU 117 confirms whether or not to initialize. When the initialization is performed (YES in S114), CPU 117 advances the process to S103. When the initialization is not performed (NO in S114), the CPU 117 ends the image processing. With the above processing, it is possible to set the range in which the first recognition processing and the second recognition processing are performed on the image obtained by the camera 110 installed on the ceiling.

On the other hand, in S115, the CPU 117 sets the installation location on the horizontal wall. Next, in S116, the CPU 117 sets shooting parameters so that an optimum image corresponding to the installed shooting environment can be obtained. Next, in S117, the CPU 117 confirms whether or not to perform the third recognition process based on the instruction received from the management device 120. CPU117 advances a process to S118, when performing the 3rd recognition process (it is YES at S117). When the third recognition process is not performed (NO in S117), CPU 117 advances the process to S120.

In S118, the CPU 117 specifies the third recognition process based on the installation position (horizontal wall) set in S115. Then, the CPU 117 specifies the third reference range corresponding to the third recognition process as the recognition range of the recognition unit 115 in the video. Then, the CPU 117 controls the display unit 130 to display the third reference range via the management device 120. The processing of S107, S110, and S118 is an example of display processing. Next, in S119, the CPU 117 sets a third application range to which the third recognition process is applied, according to the operation of the setting frame by the user. Here, the third applicable range is a range set according to a user operation as a range in which the third recognition process is performed. In this way, the user can set a desired range as the third applicable range within the third reference range.

As described above, the CPU 117 of the present embodiment determines the reference range based on the installation location of the camera 101. However, instead of the installation location, the reference range may be determined based on the shooting direction of the camera 101. In this case, the CPU 117 receives the installation information indicating the back photographing direction. Then, for example, when the shooting direction is downward, the reference range is determined as in the case where the installation place is the ceiling. On the other hand, for example, when the shooting direction is the horizontal direction, the reference range is determined as in the case where the installation place is the horizontal wall. In this way, the CPU 117 of the present embodiment can determine the reference range based on the installation information regarding the installation of the camera 101.

Next, in S120, the CPU 117 confirms whether or not there is a change in the set applicable range. When there is no change (YES in S120), CPU 117 ends the image processing. If there is a change (NO in S120), CPU 117 advances the process to S121. In S121, the CPU 117 confirms whether or not to reset the first applicable range. When resetting (YES in S121), CPU 117 advances the process to S117. CPU117 advances a process to S114, when not resetting (it is NO at S121).

FIG. 4 is a diagram showing a display example of the display unit 130 when the camera is installed on the ceiling. A real-time video (image) 410 obtained by the camera 110 is displayed on the screen 400 shown in FIG. The image 410 is the same image as the image 200. The screen 400 further displays selection buttons 401 and 402 for selecting the installation location. When the user selects the selection buttons 401 and 402 on the screen 400, the management device 120 accepts the installation location information and transmits it to the camera 110. Since the camera 110 is installed on the ceiling, the user selects the ceiling selection button 401. As a result, the image processing proceeds to S104. The screen operation is configured to be performed by a click operation with a mouse of a personal computer (PC) or a tap operation with a tablet terminal or the like, but it may be set using other means.

Then, when the installation place is set, a broken line 420 indicating the first reference range is displayed in the image 410 as shown in FIG. 4B. In the first reference range, the user clicks the rectangular two points 421 and 422 to be the first applicable range, and then presses the OK button 403 if there is no problem. As a result, the solid line 423 indicating the first applicable range is displayed on the display unit 130, and the first applicable range is set in the camera 110. On the other hand, when the first recognition processing for the central area is unnecessary, the user presses the OK button 403 without designating the first applicable range. As a result, in image processing, NO is determined in S106, the first applicable range is not set, and the process proceeds to S109.

Subsequently, as shown in FIG. 4C, a broken line 430 indicating the second reference range is displayed on the image 410. The outside of the broken line 430 is the second reference range. The user clicks the point 431 within the second reference range, and if there is no problem, presses the OK button 403. As a result, the solid line 432 indicating the second applicable range is displayed on the display unit 130, and the second applicable range is set in the camera 110. When the second recognition process for the outer area is unnecessary, the user presses the OK button 403 without designating the second applicable range. As a result, in image processing, NO is determined in S109, the second applicable range is not set, and the process proceeds to S112. In the present embodiment, the case of setting only the left range of the outer region has been described as an example, but the second applicable range can be similarly set for the right side, the upper side, and the left side.

Then, as shown in FIG. 4D, a reset button 404 is displayed on the screen 400. When the reset button 404 is selected, an OK button and a NO button are displayed together with an inquiry "Do you want to initialize?". If the OK button is selected, in image processing, YES is determined in S114, and the process returns to S106 again. If the NO button is selected, in image processing, it is determined as NO in S114, and the image processing ends.

As described above, when the installation place is on the ceiling, the camera 110 can display the reference range suitable for each of the two recognition processes. This allows the user to specify a range suitable for recognition as the applicable range of each recognition process.

FIG. 5 is a diagram showing a display example of the display unit 130 when the installation place of the camera 110 is a horizontal wall. A real-time image (image) 510 obtained by the camera 110 is displayed on the screen 500 shown in FIG. The image 510 is the same image as the image 230. The upper side of the curve 511 corresponds to the ceiling, the curves 511 to 512 correspond to the wall, and the curves 512 to 513 correspond to the floor. The screen 500 further displays setting buttons 501 and 502 for setting the installation location. Since the camera 110 is installed on the horizontal wall, the user selects the selection button 502 on the horizontal wall. As a result, the image processing proceeds to S115.

Next, as shown in FIG. 5B, a broken line 520 indicating the third reference range is displayed on the image 510. Since the upper part of the image 510 is the ceiling and the lower part is the lateral wall, the range excluding these regions is defined as the third reference range. Within the third reference range, the user clicks two rectangular points 521 and 522 that should be the third applicable range, and if there is no problem, presses the OK button 503. As a result, the solid line 523 indicating the third applicable range is displayed on the display unit 130, and the third applicable range is set in the camera 110. When the third recognition process for the central area is unnecessary, the user presses the OK button 503 without designating the third applicable range. As a result, in the image processing, it is determined as NO in S117, the third applicable range is not set, and the processing proceeds to S120.

Then, as shown in FIG. 5C, a reset button 504 is displayed on the screen 500. When the reset button 504 is selected, an OK button and a NO button are displayed together with an inquiry “Do you want to initialize?”. If the OK button is selected, in image processing, YES is determined in S114, and the process returns to S106 again. If the NO button is selected, in image processing, it is determined as NO in S114, and the image processing ends.

As described above, when the installation place is the horizontal wall, no person exists in the upper and lower regions of the image 510. On the other hand, in the camera 110 of the present embodiment, since the areas excluding these areas are set as the third reference range, it is possible to avoid setting these areas as the object recognition range.

As described above, in the camera system 100 according to the first embodiment, in a distorted image, the recognition accuracy of a specific subject is improved by limiting the range to which the recognition process is applied. You can

As a first modification of the first embodiment, the image processing described in the present embodiment may be executed by one or more devices of the camera system 100, and its execution subject is limited to the camera 110. Not a thing. For example, the management device 120 may include the video processing unit 113, the expansion unit 114, and the recognition unit 115, and the management device 120 may perform image processing.

A second modification will be described. In the present embodiment, the case has been described as an example where two locations, the ceiling and the lateral wall, can be set as the installation locations, but the number and types of installation locations are not limited to the embodiments. It is assumed that the camera 110 is preset with a recognition process corresponding to each installation place that can be set as an installation place and a reference range corresponding to the recognition process.

(Second embodiment)
FIG. 6 is an overall view of a camera system 600 according to the second embodiment. Here, the differences between the camera system 600 according to the second embodiment and the camera system 100 according to the first embodiment will be described. The camera system 600 includes a camera 601, which is an imaging device, a management device 120, and a display unit 130. The camera 601 according to the second embodiment has two image processing units 610 and 620.

In the camera 601, the image obtained by the image pickup unit 112 is input to the distribution unit 602. The distribution unit 602 divides the image into two, and distributes one to the first image processing unit 610 and the other to the second image processing unit 620 and transmits the divided image. The first image processing unit 610 and the second image processing unit 620 are, for example, video processing LSIs for processing video data.

The first image processing unit 610 includes a video processing unit 611, a development unit 612, a recognition unit 613, a network communication unit 614, an inter-chip communication unit 615, a CPU 616, and a storage unit 617. There is. The video processing unit 611, the expansion unit 612, the recognition unit 613, the network communication unit 614, the CPU 616, and the storage unit 617 are the video processing unit 113, the expansion unit 114, the recognition unit 115, and the communication unit 116, respectively, according to the first embodiment. It is similar to the CPU 117 and the storage unit 118. The inter-chip communication unit 615 performs data communication between the first image processing unit 610 and the second image processing unit 620.

The second image processing unit 620 includes a video processing unit 621, a development unit 622, a recognition unit 623, a network communication unit 624, an inter-chip communication unit 625, a CPU 626, and a storage unit 627. There is. The video processing unit 621, the expansion unit 622, and the recognition unit 623 are similar to the video processing unit 611, the expansion unit 612, and the recognition unit 613, respectively. The network communication unit 624, the inter-chip communication unit 625, the CPU 626, and the storage unit 627 are similar to the network communication unit 614, the inter-chip communication unit 615, the CPU 616, and the storage unit 617, respectively.

In the present embodiment, the camera 601 has a configuration in which one channel needs to communicate with the external network. For this reason, it is assumed that only the network communication unit 614 performs communication via the network 150. When the camera 601 wants to perform wideband data communication with another device, each of the network communication unit 614 and the network communication unit 624 may perform communication via the network 150.

FIG. 7 is a flowchart showing image processing by the camera 601 according to the second embodiment. Normally, when processing high-resolution video, one processing block resource can handle only a predetermined processing amount. In this case, the first image processing unit 610 performs processing using a plurality of processing block resources. The image processing is processing using a plurality of block resources. First, after the initialization, the distribution unit 602 simply divides the image into a plurality of fixed areas, and the image processing units 610 and 620 input the image data for each divided area. It should be noted that among the respective processes shown in FIG. 7, the same processes as the respective processes of the image processing in the first embodiment are denoted by the same reference numerals.

After the processing of S101, the CPU 117 advances the processing to S700. In S700, the CPU 616 controls the lens 111 and the image capturing unit 112 to start shooting. Then, the distribution unit 602 divides the image into two vertically. The number of divisions and division positions are not limited to those in the embodiment. Next, in S701, the first image processing unit 610 and the second image processing unit 620 perform image processing on the images respectively supplied. Next, in step S<b>702, the first image processing unit 610 receives the image-processed video from the second image processing unit 620, and integrates the image-processed video and the received video in the own device. Next, in step S703, the network communication unit 614 distributes the integrated video to the management device 120. When the management device 120 receives the video, the management device 120 displays the video on the display unit 130. Next, the CPU 616 advances the process to S103.

In S103, the CPU 616 confirms whether the installation location of the camera 110 is the ceiling or the lateral wall. When the installation place is the ceiling (YES in S103), CPU 616 advances the process to S104. When the installation location is on the horizontal wall (NO in S103), CPU 616 advances the process to S115. In S104, the CPU 616 sets the installation place to the ceiling. Next, in S105, the CPU 616 sets shooting parameters so that an optimum image corresponding to the installed shooting environment can be obtained.

Next, in S106, the CPU 616 confirms whether to perform the first recognition process. When performing the first recognition process (YES in S106), CPU 616 advances the process to S107. When the first recognition process is not performed (NO in S106), CPU 616 advances the process to S109. After the processing of S107 and S108, the CPU 616 advances the processing to S704.

In S704, the CPU 616 changes the division position so that the division position is not included in the first applicable range. This process is an example of a determination process for determining the division position. The CPU 616 also considers the processing capabilities of the first image processing unit 610 and the second image processing unit 620 when changing the division position.

The CPU 616 then advances the process to S109. In S109, the CPU 616 confirms whether or not to perform the second recognition process based on the instruction received from the management device 120. When performing the second recognition process (YES in S109), CPU 616 advances the process to S110. When the second recognition process is not performed (NO in S109), CPU 616 advances the process to S112.

In S110, the CPU 616 controls to display the second reference range associated with the second recognition process on the display unit 130 via the management device 120. Next, in S111, the CPU 616 sets a second application range to which the second recognition process is applied, in accordance with the operation of the setting frame by the user. The CPU 616 then advances the process to S705. In S705, the CPU 616 further considers the second applicable range with respect to the area other than the range divided in S704, and further changes the dividing position so that the video processing performance is maximized. The CPU 616 then advances the process to S112.

Further, after the processing of S119, the CPU 616 advances the processing to S706. In S706, the CPU 616 changes the division position so that the division position is not included in the third applicable range. The CPU 616 then advances the process to S120.

As described above, the camera 601 according to the second embodiment performs processing so that the divisional position is not included in the image to be subjected to the recognition process with respect to the high-resolution image, so that the recognition accuracy and the image processing can be improved. The efficiency of can be improved.

FIG. 8 is a diagram showing a display example of the display unit 130 when the camera is installed on the ceiling. Among the image processing according to the second embodiment, differences from the image processing according to the first embodiment will be described with reference to FIG. 8. A screen 800 shown in FIG. 8A shows a real-time video (image) 810 obtained by the camera 110. The image 810 is the same image as the image 200. The screen 800 further displays setting location selection buttons 801 and 802. The broken line 811 in the figure is a line indicating the dividing position of the image, but the dividing position is not displayed on the actual screen 800.

Then, when the installation position is set, a broken line 820 indicating the first reference range is displayed in the image 410, as shown in FIG. 8B. Then, when the points 821 and 822 are specified by the user operation, the first application range of the rectangle having these points as vertices is set, and the solid line 823 indicating the first application range is displayed. At this time, the CPU 616 changes the division position so that the division position is not included in the first applicable range. In the present embodiment, the CPU 616 changes the division position by setting the upper limit of the video size (area) in charge of the second image processing unit 620 to 1/2 of the size of the entire video.

Subsequently, as shown in FIG. 8C, a broken line 830 indicating the second reference range is displayed on the image 810. The outside of the broken line 830 is the second reference range. The user clicks the point 831 within the second reference range and presses the OK button 803 if there is no problem. As a result, the solid line 832 indicating the second applicable range is displayed on the display unit 130, and the second applicable range is set in the camera 110.

The CPU 616 calculates the data amount of the second applicable range according to the data amount of the first applicable range, calculates the area of the area outside the arc passing through the point 831, and then determines the processable range. Is displayed. In the present embodiment, the CPU 616 calculates the area along the counterclockwise arc from the point 831. However, the CPU 616 may calculate the area along the clockwise arc from the point 831.

Further, the CPU 616 changes the division position so that the division position is not included in the first application range and the second application range. The other configurations and processes of the camera system 600 according to the second embodiment are the same as the configurations and processes of the camera system 100 according to the first embodiment.

As described above, in the camera system 600 according to the second embodiment, when the video is distributed to the two image processing units, the video is divided into the first application range and the second application range. Divide the image so that the position is not included. Thereby, the processing efficiency can be improved.

As described above, according to each of the above-described embodiments, it is possible to improve the recognition accuracy of a specific subject in an image with distortion.

Although the preferred embodiments of the present invention have been described above in detail, the present invention is not limited to the specific embodiments, and various modifications are possible within the scope of the gist of the present invention described in the claims. -Can be changed.

(Other embodiments)
The present invention supplies a program that implements one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program. It can also be realized by the processing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

100 camera system 110 camera 120 management device 130 display device

Claims (11)

Designation means for designating an application range in which a specific subject is recognized from an image obtained by the image pickup means,
An image processing apparatus comprising: a display processing unit that controls the display unit to display a reference range in which the applicable range can be specified by the specifying unit. The image processing apparatus according to claim 1, wherein the designation unit designates the applicable range according to a user operation. Further comprising specifying means for specifying the reference range based on installation information of the image pickup means,
The image processing apparatus according to claim 1, wherein the display processing unit controls to display the reference range specified by the specifying unit. The image processing apparatus according to claim 3, wherein the specifying unit specifies the reference range based on installation information input according to a user operation. The specifying unit specifies the first reference range as a range in which the first recognition processing is possible when the installation place indicated by the installation information is a ceiling, and further, the second reference different from the first reference range. The image processing apparatus according to claim 3, wherein the reference range is specified as a range in which the second recognition processing different from the first recognition processing is possible. When the installation location indicated in the installation information is a horizontal wall, the identifying unit defines a third reference range different from the first reference range and the second reference range as the first recognition process and the first reference range. The image processing apparatus according to claim 5, wherein the image processing apparatus is specified as a range in which a third recognition process different from the second recognition process can be performed. Determining means for determining the division position of the image based on the reference range;
7. The distribution device further comprising: a distribution unit that divides the image obtained by the image pickup unit at the division position determined by the determination unit and distributes the image to a plurality of image processing units. The image processing apparatus according to item 1. The image processing apparatus according to claim 1, wherein the reference range is a preset range. 9. The image processing apparatus according to claim 1, further comprising a recognition unit that executes recognition processing of a specific subject with respect to the applicable range. An image processing method executed by an image processing apparatus, comprising:
A designating step of designating an applicable range in which a specific subject is recognized from the image obtained by the image pickup means;
And a display processing step of controlling the display means to display a reference range capable of designating the applicable range in the designating step. A program for causing a computer to operate as each unit of the image processing apparatus according to claim 1.

Images