JP7638191B2 - Imaging device, image analysis system, encoding method, and encoding program - Google Patents

Description

The present disclosure relates to an imaging device, an image analysis system, an encoding method, and an encoding program.

Users can view video captured by surveillance cameras and other devices in real time. Users can also view the video recorded on a recording medium. In recent years, computer-based video analysis technology has become widespread. Using this technology, people, types of objects, object attributes, and so on can be identified.

When analyzing one image from a video (i.e., multiple images), the analysis is not performed on the entire image. For example, when performing face recognition, a face area is first detected from within the image. After the face area is detected, face recognition is performed on the face area. The load of face area detection processing is small. Therefore, face area detection processing may be performed on an imaging device. On the other hand, the load of face recognition is large. Therefore, face recognition is performed on an analysis device. Here, a technology has been proposed that performs intra-coding on face areas (see Patent Document 1).

JP 2017-225068 A

As with the above technology, one possible method is to treat images that include face areas as intraframes. If a person barely moves, almost all of the image will be intraframes. Or more precisely, if the subject barely moves, almost all of the image will be intraframes. Having almost all images be intraframes means that the amount of data sent to downstream devices will be large.

The purpose of this disclosure is to reduce the amount of data transmitted.

An imaging device according to one aspect of the present disclosure is provided. The imaging device captures an image of an object. The imaging device includes an acquisition unit that acquires a first image including the object, and acquires information indicating the importance of a second image that is an image generated before the first image and includes the object, a detection position of the object in the second image, and a second detection area of the object in the second image, and a detection unit that detects the object based on the first image, and compares the detection position of the object in the second image with the detection position of the object in the first image, compares the second detection area with the first detection area of the object in the first image, and detects whether the difference in detection position and the difference in detection area are equal to or less than a predetermined first threshold. and if the first detection area is narrower than the second detection area or the first detection area is the same as the second detection area, the importance determination unit determines the importance of the first image to be a value smaller than the importance of the second image; if the importance of the first image is smaller than a predetermined second threshold, the encoding determination unit determines the first image to be a frame other than an intraframe; if the determination indicates a frame other than an intraframe, the encoding unit encodes the first image as a frame other than an intraframe; and a transmission unit transmits the encoded data.

This disclosure makes it possible to reduce the amount of data transmitted.

FIG. 1 illustrates an image analysis system. FIG. 2 is a diagram illustrating hardware included in the imaging apparatus. FIG. 2 is a block diagram showing functions of the imaging apparatus. 10 is a flowchart illustrating an example of processing executed by the imaging device. FIG. 13 is a diagram illustrating an example of metadata. 11 is a flowchart (part 1) illustrating an example of an importance determination process. 13 is a flowchart (part 2) illustrating an example of the importance determination process. 13A and 13B are diagrams illustrating a specific example of importance determination processing. FIG. 13 is a flowchart illustrating an example of an intra-coding decision process. FIG. 2 is a block diagram showing the functions of a recording device and an analysis device. FIG. 13 is a diagram showing an example of processed metadata. 4 is a flowchart showing a part of the process of the recording device. 13 is a flowchart illustrating an example of a metadata transmission process. 13 is a flowchart illustrating an example of a transmission process of encoded data. 13 is a flowchart illustrating an example of a process executed by the analysis device.

The following describes an embodiment with reference to the drawings. The following embodiment is merely an example, and various modifications are possible within the scope of this disclosure.

Embodiment
1 is a diagram showing an image analysis system. The image analysis system includes an imaging device 100, a recording device 200, and an analysis device 300. The imaging device 100 and the recording device 200 are connected via a network 10. The recording device 200 and the analysis device 300 are connected via a network 20.

The imaging device 100 captures an image of an object. The object may be a person, an animal, a machine (e.g., an automobile), etc. The imaging device 100 is also a device that executes an encoding method.
The recording device 200 stores the data received from the imaging device 100 .
Analysis device 300 performs analysis using the data stored in recording device 200. For example, analysis device 300 is a PC (Personal Computer).

Next, the hardware of the imaging device 100 will be described.
2 is a diagram showing hardware included in the image capturing apparatus 100. The image capturing apparatus 100 includes a processor 101, a volatile storage device 102, and a non-volatile storage device 103.

The processor 101 controls the entire imaging device 100. For example, the processor 101 is a CPU (Central Processing Unit) or an FPGA (Field Programmable Gate Array). The processor 101 may be a multiprocessor. The imaging device 100 may also have a processing circuit.

The volatile storage device 102 is a main storage device of the imaging device 100. For example, the volatile storage device 102 is a random access memory (RAM). The non-volatile storage device 103 is an auxiliary storage device of the imaging device 100. For example, the non-volatile storage device 103 is a hard disk drive (HDD) or a solid state drive (SSD).
Similar to the imaging device 100, the recording device 200 and the analysis device 300 each include a processor, a volatile storage device, and a non-volatile storage device.

Next, functions of the imaging device 100 will be described.
3 is a block diagram showing the functions of the imaging apparatus 100. The imaging apparatus 100 includes an imaging unit 110, an acquisition unit 120, a detection unit 130, an importance determination unit 140, an encoding determination unit 150, an encoding unit 160, a metadata generation unit 170, a multiplexing unit 180, and a transmission unit 190.

The acquisition unit 120, the detection unit 130, the importance determination unit 140, the encoding decision unit 150, the encoding unit 160, the metadata generation unit 170, the multiplexing unit 180, and the transmission unit 190 may be partly or entirely realized by a processing circuit. In addition, the acquisition unit 120, the detection unit 130, the importance determination unit 140, the encoding decision unit 150, the encoding unit 160, the metadata generation unit 170, the multiplexing unit 180, and the transmission unit 190 may be partly or entirely realized as a program module executed by the processor 101. For example, the program executed by the processor 101 is also referred to as an encoding program. For example, the encoding program is recorded on a recording medium.

The imaging unit 110 captures an image of an object.
The acquisition unit 120 acquires an image. For example, the acquisition unit 120 acquires an image including an object. Also, the acquisition unit 120 acquires a video (i.e., a plurality of images). Therefore, this sentence can be expressed as follows. The acquisition unit 120 sequentially acquires images from the imaging unit 110. Here, an image is also called a frame.
As described above, the acquisition unit 120 acquires an image. For example, the acquired image is also referred to as a currently generated image or a first image.

The acquisition unit 120 acquires information indicating the importance of a second image, which is an image generated before the first image and includes an object. The second image may be an image generated one image before the first image, or an image generated two images before the first image. The acquisition unit 120 also acquires information indicating the detection position of the object in the second image and the detection area of the object in the second image. In the following description, the second image is referred to as the previously generated image.

The functions of the detection unit 130, importance determination unit 140, encoding determination unit 150, and encoding unit 160 will be described later.
The metadata generating unit 170 generates metadata for the acquired image.
The multiplexing unit 180 combines the encoded data obtained by the encoding unit 160 and the metadata into one to generate multiplexed data. For example, the multiplexed data may be called first multiplexed data.
The transmitting unit 190 transmits the multiplexed data to the recording device 200 .

Next, the process executed by the image capture device 100 will be described with reference to a flowchart.
FIG. 4 is a flowchart illustrating an example of processing executed by the imaging device.
(Step S11) The acquisition unit 120 acquires an image from the imaging unit 110. Note that the image is the currently generated image or the first image.
(Step S12) The detection unit 130 executes an object detection process based on the acquired image. For example, the detection unit 130 executes the object detection process using an object recognition technique. If an object is included in the image, the detection unit 130 detects the object.

(Step S13) The importance determining unit 140 executes an importance determining process.
(Step S14) The coding decision unit 150 performs intra-coding decision processing.

(Step S15) The encoding unit 160 performs encoding according to the determined content. In particular, if the determined content indicates an intra frame, the encoding unit 160 encodes the image acquired in step S11 as an intra frame. Furthermore, if the determined content indicates a frame other than an intra frame, the encoding unit 160 encodes the image acquired in step S11 as a P frame (Predicted Frame) or a B frame (Bi-directional Predicted Frame).

(Step S16) The metadata generation unit 170 generates metadata based on the detection result by the detection unit 130 and the importance determined by the importance determination process. Here, examples of metadata are shown.

Figure 5 shows an example of metadata. Metadata is used as auxiliary information for image analysis. Metadata includes object information and image information. For example, object information is information based on detection results. Image information includes information indicating whether it is an intraframe, importance, etc.

(Step S17) The multiplexing unit 180 generates multiplexed data by combining the encoded data obtained by the encoding unit 160 and the metadata. The multiplexing unit 180 may generate multiplexed data including the image capture time.
(Step S18) The transmitting unit 190 transmits the multiplexed data to the recording device 200. Alternatively, the transmitting unit 190 may transmit only the encoded data to the recording device 200. Furthermore, the transmitting unit 190 may transmit the multiplexed data or the encoded data to a device other than the recording device 200.

Next, the importance determination process will be described.
6 is a flowchart (part 1) showing an example of the importance determination process. The process in FIGS. 6 and 7 corresponds to step S13.

Here, importance will be explained. Importance is expressed as a value between 0 and 100. The closer the value is to 0, the lower the importance. The closer the value is to 100, the higher the importance. The initial value is 50. Also, importance does not have to be expressed in the range between 0 and 100. In other words, the expression of importance is not limited to this range.

(Step S21) The importance determination unit 140 determines whether or not an object has been detected in the object detection process. If an object has been detected, the importance determination unit 140 assigns an identification number to the image generated this time (i.e., the image acquired in step S11). Then, the process proceeds to step S22. If an object has not been detected, the importance determination unit 140 determines the importance of the image generated this time to be 0. Then, the process ends.

(Step S22) The importance determination unit 140 determines whether or not an object has been detected from the previously generated image. If an object has not been detected from the previously generated image, the importance determination unit 140 determines that a new object has been detected. Then, the process proceeds to step S26. If an object has been detected from the previously generated image, the process proceeds to step S23.

(Step S23) The importance determination unit 140 compares the detection position of the object in the previously generated image with the detection position of the object in the currently generated image. The importance determination unit 140 also compares the detection area of the object in the previously generated image (e.g., also referred to as the second detection area) with the detection area of the object in the currently generated image (e.g., also referred to as the first detection area).

(Step S24) The importance determination unit 140 judges whether the difference in the detection position and the difference in the detection area are equal to or less than a predetermined threshold. Here, the threshold is also referred to as a first threshold. If the difference in the detection position and the difference in the detection area are equal to or less than the threshold, the importance determination unit 140 judges that the object in the image generated previously and the object in the image generated this time are the same object. Then, the process proceeds to step S25. If the difference in the detection position and the difference in the detection area are not equal to or less than the threshold, the process proceeds to step S32.
The threshold value to be compared with the difference in the detection position and the threshold value to be compared with the difference in the detection area may be different.

(Step S25) The importance determining unit 140 changes the identification number assigned to the currently generated image to the identification number assigned to the previously generated image, and the process then proceeds to step S31.
(Step S26) The importance determining unit 140 determines the importance of the currently generated image to be an initial value, and the process then ends.

FIG. 7 is a flowchart (part 2) illustrating an example of the importance determination process.
(Step S31) The importance determining unit 140 determines whether the detection area of the image generated this time (hereinafter, "current detection area") is larger than the detection area of the image generated previously (hereinafter, "previous detection area").
If the current detection area is larger than the previous detection area, the process proceeds to step S32. If the current detection area is smaller than the previous detection area, or if the current detection area and the previous detection area are the same, the process proceeds to step S33.

(Step S32) The importance determining unit 140 determines the importance of the currently generated image to be greater than the importance of the previously generated image, and the process then ends.
(Step S33) The importance determining unit 140 determines the importance of the currently generated image to be a value smaller than the importance of the previously generated image, and the process then ends.
When step S32 is executed after step S24, the importance determining section 140 may determine the importance of the currently generated image to be an initial value.

Here, a specific example of the importance determination process will be shown.
8A and 8B are diagrams showing a specific example of the importance determination process. Fig. 8A shows a case where step S32 is executed after step S31. Fig. 8B shows a case where step S33 is executed after step S31.

An image 400 at time t-n shown in Figures 8(A) and (B) is assumed to be a previously generated image. Note that n is a positive integer. Figures 8(A) and (B) also show a previously detected area 401. The importance of the previously generated image is 40.
An image 410 at time t shown in Figures 8A and 8B is an image generated this time. Also, Figures 8A and 8B show a current detection area 411.

8A, current detection area 411 is larger than previous detection area 401. Therefore, importance determination unit 140 determines the importance of image 410 to be 70. Furthermore, the larger the detection area, the higher the analysis accuracy of analysis device 300. Therefore, when the detection area is large, a large importance value is determined.
8B, the current detection area 411 is narrower than the previous detection area 401. Therefore, the importance determining unit 140 determines the importance of the image 410 to be 20.

As shown in Figure 7, if the object is barely moving, the importance of the image gradually decreases. Also, if the object is cut off, or if the object is covered by an obstacle and part of the object is missing, a small importance value is determined.

The importance may also be determined by the analysis content of the analysis device 300. For example, when the analysis device 300 analyzes the degree of congestion, the importance determination unit 140 determines a large value of importance for an image with a large number of faces. The importance determination unit 140 determines a small value of importance for an image with a small number of faces. Also, for example, when the analysis device 300 analyzes an image in which an object is large, the importance determination unit 140 compares a predetermined reference range with the current detection area. If the current detection area is wider than the reference range, the importance determination unit 140 determines the importance of the image generated this time to a large value. If the current detection area is narrower than the reference range, the importance determination unit 140 determines the importance of the image generated this time to a small value. Also, for example, when the analysis device 300 analyzes an image generated at a specific time, the importance determination unit 140 determines the importance of the image generated at the specific time to a large value. The importance determination unit 140 then determines the importance of images generated at other times to be a smaller value.

Next, the intra-coding decision process will be described.
9 is a flowchart illustrating an example of the intra-coding decision process, which corresponds to step S14.
(Step S41) The encoding decision unit 150 compares the importance of the currently generated image with a predetermined threshold. For example, the threshold is 50, 60, etc. Note that the threshold may be changed depending on the detected scene and object. The threshold is also called the second threshold.

(Step S42) The encoding decision unit 150 determines whether the importance of the currently generated image is equal to or greater than the threshold. If the importance of the currently generated image is equal to or greater than the threshold, the process proceeds to step S43. If the importance of the currently generated image is less than the threshold, the encoding decision unit 150 decides not to make the currently generated image an intraframe. That is, the encoding decision unit 150 decides that the currently generated image will be a frame other than an intraframe. In detail, the encoding decision unit 150 decides that the currently generated image will be a P frame or a B frame. Then, the process ends.
(Step S43) The encoding decision unit 150 decides that the currently generated image is an intra-frame.

Here, if the subject is barely moving, almost all images may be set as intraframes. If almost all images are intraframes, the amount of data sent to downstream devices (e.g., recording devices) will be large.

According to the embodiment, when the object hardly moves, the importance is reduced. The reduced importance increases the number of times the currently generated image is determined to be a P frame or a B frame. In other words, the number of times the currently generated image is determined to be an intra frame decreases. Therefore, the amount of data transmitted when few intra frames and many P frames or B frames are determined is less than the amount of data transmitted when almost all images are determined to be intra frames. Therefore, the imaging device 100 or the image analysis system can reduce the amount of data transmitted.

Next, the recording device 200 and the analysis device 300 will be described.
10 is a block diagram showing the functions of the recording device and the analysis device. The recording device 200 has a receiving unit 210, a separating unit 220, an encoded data storage unit 230, a processing unit 240, a metadata storage unit 250, a metadata transmission control unit 260, and an encoded data transmission control unit 270.

The encoded data storage unit 230 and the metadata storage unit 250 may be realized as storage areas secured in a volatile storage device or a non-volatile storage device included in the recording device 200 .
Some or all of the receiving unit 210, the separating unit 220, the processing unit 240, the metadata transmission control unit 260, and the encoded data transmission control unit 270 may be realized by a processing circuit included in the recording device 200. Also, some or all of the receiving unit 210, the separating unit 220, the processing unit 240, the metadata transmission control unit 260, and the encoded data transmission control unit 270 may be realized as program modules executed by a processor included in the recording device 200.

The analysis device 300 has a metadata communication unit 310, a memory unit 320, an encoded data control unit 330, an encoded data communication unit 340, a decoding unit 350, and an analysis unit 360.

The storage unit 320 may be realized as a storage area reserved in a volatile storage device or a non-volatile storage device included in the analysis device 300 .
Some or all of the metadata communication unit 310, the encoded data control unit 330, the encoded data communication unit 340, the decoding unit 350, and the analysis unit 360 may be realized by a processing circuit included in the analysis device 300. Some or all of the metadata communication unit 310, the encoded data control unit 330, the encoded data communication unit 340, the decoding unit 350, and the analysis unit 360 may be realized as program modules executed by a processor included in the analysis device 300.

First, the function of the recording device 200 will be described.
The receiving unit 210 receives the multiplexed data. The separating unit 220 separates the multiplexed data into encoded data and metadata. The separating unit 220 stores the encoded data in the encoded data storage unit 230.
The processing unit 240 processes the metadata. An example of the processed metadata is shown below.

11 is a diagram showing an example of processed metadata, in which time information and encoding information are added.
When the multiplexed data includes a shooting time, the processing unit 240 adds the shooting time to the metadata as time information. When the multiplexed data does not include a shooting time, the processing unit 240 adds the time at which the multiplexed data was received to the metadata as a shooting time.
The processing unit 240 adds, as encoding information, the encoded data number corresponding to the encoded data included in the multiplexed data to the metadata.
The processing unit 240 stores the processed metadata in the metadata storage unit 250 .

When the metadata transmission control unit 260 receives a transmission request from the analysis device 300 , it transmits the metadata stored in the metadata storage unit 250 to the analysis device 300 .
When the encoded data transmission control unit 270 receives a transmission request from the analysis device 300 , it transmits the encoded data to the analysis device 300 .

Next, the function of the analysis device 300 will be described.
The metadata communication unit 310 transmits a request to transmit metadata to the recording device 200. When the metadata communication unit 310 receives metadata, it stores the metadata in the storage unit 320.

The encoded data control unit 330 acquires analysis condition information indicating the analysis conditions. For example, the encoded data control unit 330 acquires the analysis condition information from the storage unit 320. Also, for example, the encoded data control unit 330 acquires the analysis condition information input by the user. For example, the analysis condition indicates an intraframe with an importance level of 70 or higher. Also, the analysis condition may include the imaging time or imaging duration. For example, the imaging duration is 12:00 to 13:00.

The encoded data control unit 330 identifies the encoded data to be analyzed based on the analysis condition information and the metadata. In detail, the encoded data control unit 330 identifies the encoded data number of the encoded data to be analyzed.

The encoded data communication unit 340 transmits a transmission request for the identified encoded data (i.e., the encoded data to be analyzed) to the recording device 200. The transmission request includes the identified encoded data number. The encoded data communication unit 340 then receives the encoded data to be analyzed.

The decoding unit 350 decodes the received encoded data and obtains metadata corresponding to the received encoded data from the storage unit 320.
The analysis unit 360 performs analysis based on the decoded data (i.e., image) and metadata.

Next, the processing of the recording device 200 will be described with reference to a flowchart.
FIG. 12 is a flowchart showing a part of the process of the recording device.
(Step S51) The receiving unit 210 receives multiplexed data.
(Step S52) The demultiplexer 220 demultiplexes the multiplexed data into encoded data and metadata.
(Step S<b>53 ) The separator 220 stores the coded data in the coded data storage unit 230 .

(Step S54) The processing unit 240 processes the metadata.
(Step S55) The processing unit 240 stores the processed metadata in the metadata storage unit 250.
Note that step S53 may be performed simultaneously with steps S54 and S55. Also, step S53 may be performed after steps S54 and S55.

FIG. 13 is a flowchart illustrating an example of a metadata transmission process.
(Step S61) The metadata transmission control unit 260 determines whether or not a metadata transmission request has been received from the analysis device 300. If a metadata transmission request has been received, the process proceeds to step S62. If a metadata transmission request has not been received, the process ends.
(Step S62) The metadata transmission control unit 260 acquires metadata from the metadata storage unit 250.
(Step S63) The metadata transmission control unit 260 transmits the metadata to the analysis device 300.

FIG. 14 is a flowchart illustrating an example of a transmission process of encoded data.
(Step S71) The encoded data transmission control unit 270 determines whether or not a request to transmit encoded data has been received from the analysis device 300. If a request to transmit encoded data has been received, the process proceeds to step S72. If a request to transmit encoded data has not been received, the process ends.
(Step S72) The encoded data transmission control unit 270 obtains the encoded data from the encoded data storage unit 230.
(Step S73) The encoded data transmission control unit 270 transmits the encoded data to the analysis device 300.

Next, the processing of analysis device 300 will be described with reference to a flowchart.
FIG. 15 is a flowchart illustrating an example of a process executed by the analysis device.
(Step S81) The metadata communication unit 310 transmits a metadata transmission request to the recording device 200.
(Step S82) The metadata communication unit 310 determines whether or not metadata has been received. If metadata has been received, the process proceeds to step S83. If metadata has not been received, the metadata communication unit 310 waits until metadata is received.
(Step S83) The metadata communication unit 310 stores the metadata in the storage unit 320.

(Step S84) The encoded data control unit 330 identifies the encoded data to be analyzed from the encoding information of the metadata.
(Step S85) The encoded data communication unit 340 transmits a request to the recording device 200 to transmit the identified encoded data.
(Step S86) The encoded data communication unit 340 determines whether or not encoded data has been received. If encoded data has been received, the process proceeds to step S87. If encoded data has not been received, the encoded data communication unit 340 waits until encoded data is received.

(Step S87) The decoding unit 350 decodes the received encoded data.
(Step S88) The decoding unit 350 obtains metadata corresponding to the received encoded data from the storage unit 320.
(Step S89) The analysis unit 360 performs analysis based on the decoded data (i.e., image) and metadata.

As described above, the metadata includes the importance. Therefore, the analysis device 300 can identify images with high importance. Furthermore, the analysis device 300 does not receive all the encoded data stored in the recording device 200. The analysis device 300 receives only the identified encoded data. Therefore, the amount of communication between the recording device 200 and the analysis device 300 is reduced.

In addition, when there are many intraframes, the encoded data control unit 330 may identify the most important intraframe from the metadata. In detail, the image analysis system performs the following process. The imaging device 100 generates multiplexed data every time it generates an image, and transmits the multiplexed data to the recording device 200. Thus, the imaging device 100 transmits multiple multiplexed data to the recording device 200. The recording device 200 receives multiple multiplexed data. The recording device 200 adds an encoded data number corresponding to the encoded data to the metadata for each multiplexed data. The recording device 200 transmits the processed multiple metadata to the analysis device 300. The analysis device 300 receives the multiple metadata. The analysis device 300 detects that there are more intraframes than a predetermined threshold based on the multiple metadata. Here, the threshold is also called a third threshold. The analysis device 300 identifies the encoded data of the most important intraframe based on the multiple metadata. In particular, the analysis device 300 identifies the encoded data number of the encoded data of the most important intraframe based on multiple pieces of metadata.

The analysis device 300 may adjust the number of frames to be decoded depending on the processing load. In addition, the analysis device 300 can detect the distribution of important frames within the entire video by analyzing the metadata.

10 network, 20 network, 100 imaging device, 101 processor, 102 volatile storage device, 103 non-volatile storage device, 110 imaging unit, 120 acquisition unit, 130 detection unit, 140 importance determination unit, 150 encoding determination unit, 160 encoding unit, 170 metadata generation unit, 180 multiplexing unit, 190 transmission unit, 200 recording device, 210 reception unit, 220 separation unit, 230 encoded data storage unit, 240 processing unit, 250 metadata storage unit, 260 metadata transmission control unit, 270 encoded data transmission control unit, 300 analysis device, 310 metadata communication unit, 320 storage unit, 330 encoded data control unit, 340 encoded data communication unit, 350 decoding unit, 360 Analysis unit, 400 image, 401 previous detection area, 410 image, 411 current detection area.

Claims (9)

An imaging device for imaging an object, comprising:
an acquisition unit that acquires a first image including the object, and acquires information indicating the importance of a second image that is an image generated before the first image and includes the object, a detection position of the object in the second image, and a second detection area of the object in the second image;
A detection unit that detects the object based on the first image;
an importance determination unit that compares a detection position of the object in the second image with a detection position of the object in the first image, compares the second detection area with a first detection area of the object in the first image, and determines the importance of the first image to be a value smaller than the importance of the second image when a difference in detection position and a difference in detection area are equal to or smaller than a first threshold value and the first detection area is narrower than the second detection area or the first detection area and the second detection area are the same;
an encoding decision unit that decides that the first image is a frame other than an intra frame when the importance of the first image is smaller than a second predetermined threshold;
an encoding unit that encodes the first image as a frame other than an intraframe when the determined content indicates a frame other than an intraframe;
A transmitter that transmits the encoded data;
An imaging device having the above configuration. The importance determination unit is
When the difference between the detection positions and the difference between the detection areas are equal to or less than the first threshold value and the first detection area is larger than the second detection area, determining the importance of the first image to be greater than the importance of the second image;
The encoding determination unit is
If the importance of the first image is equal to or greater than the second threshold, determining the first image as an intra-frame;
The encoding unit is
if the determination indicates an intraframe, encoding the first image as an intraframe;
The imaging device according to claim 1 . a metadata generating unit configured to generate metadata for the first image;
The transmission unit is
transmitting multiplexed data including the encoded data and the metadata;
The imaging device according to claim 1 . An imaging device that captures an image of an object;
A recording device;
An analysis device;
Including,
The imaging device includes:
an acquisition unit that acquires a first image including the object, and acquires information indicating the importance of a second image that is an image generated before the first image and includes the object, a detection position of the object in the second image, and a second detection area of the object in the second image;
A detection unit that detects the object based on the first image;
an importance determination unit that compares a detection position of the object in the second image with a detection position of the object in the first image, compares the second detection area with a first detection area of the object in the first image, and determines the importance of the first image to be a value smaller than the importance of the second image when a difference in detection position and a difference in detection area are equal to or smaller than a first threshold value and the first detection area is narrower than the second detection area or the first detection area and the second detection area are the same;
an encoding determination unit that determines the first image to be a frame other than an intra frame when the importance of the first image is smaller than a second predetermined threshold;
an encoding unit that encodes the first image as a frame other than an intraframe when the determined content indicates a frame other than an intraframe;
A transmitter that transmits the encoded data;
having
Image analysis system. The imaging device includes:
generating metadata for the first image;
Transmitting multiplexed data including the encoded data and the metadata to the recording device;
The recording device is
receiving the multiplexed data;
adding an encoded data number corresponding to the encoded data to the metadata;
Transmitting the processed metadata to the analysis device;
The analysis device includes:
receiving the metadata;
acquiring analysis condition information indicating analysis conditions; and identifying an encoded data number of the encoded data to be analyzed based on the analysis condition information and the metadata;
Transmitting a request to transmit the encoded data to be analyzed to the recording device;
receiving the encoded data to be analyzed from the recording device;
Decoding the received encoded data;
performing an analysis based on the decrypted data and the metadata;
The image analysis system according to claim 4. The imaging device includes:
generating metadata including an importance of the first image and information indicating whether the first image is an intraframe;
Transmitting a plurality of multiplexed data including first multiplexed data which is a combination of encoded data, which is encoded data, and the metadata to the recording device;
The recording device is
receiving the plurality of multiplexed data;
adding an encoded data number corresponding to the encoded data to the metadata for each of the multiplexed data;
Transmitting the processed metadata to the analysis device;
The analysis device includes:
receiving the plurality of metadata;
Detecting whether an intraframe is present in an amount equal to or greater than a third predetermined threshold based on the plurality of metadata;
Identifying the encoded data of the most important intraframe based on the plurality of metadata;
Transmitting a transmission request for the identified encoded data to the recording device;
receiving the identified encoded data from the recording device;
Decoding the received encoded data;
performing an analysis based on the decoded data and metadata corresponding to the identified encoded data;
The image analysis system according to claim 4. the analysis device adjusts the number of frames to be decoded according to a processing load status.
The image analysis system according to any one of claims 4 to 6. An imaging device for imaging an object,
acquiring a first image including the object, and acquiring information indicating a level of importance of a second image that is an image generated before the first image and includes the object, a detection position of the object in the second image, and a second detection area of the object in the second image;
Detecting the object based on the first image;
comparing a detection position of the object in the second image with a detection position of the object in the first image, and comparing the second detection area with a first detection area of the object in the first image;
When the difference in the detection positions and the difference in the detection areas are equal to or less than a first threshold value, and the first detection area is narrower than the second detection area, or the first detection area and the second detection area are the same, the importance of the first image is determined to be a value smaller than the importance of the second image;
If the importance of the first image is less than a second predetermined threshold, the first image is determined to be a frame other than an intra frame;
If the determination indicates a non-intra frame, encoding the first image as a non-intra frame;
Transmitting the encoded data;
Encoding method. An imaging device for imaging an object,
acquiring a first image including the object, and acquiring information indicating a level of importance of a second image that is an image generated before the first image and includes the object, a detection position of the object in the second image, and a second detection area of the object in the second image;
Detecting the object based on the first image;
comparing a detection position of the object in the second image with a detection position of the object in the first image, and comparing the second detection area with a first detection area of the object in the first image;
When the difference in the detection positions and the difference in the detection areas are equal to or less than a first threshold value, and the first detection area is narrower than the second detection area, or the first detection area and the second detection area are the same, the importance of the first image is determined to be a value smaller than the importance of the second image;
If the importance of the first image is less than a second predetermined threshold, the first image is determined to be a frame other than an intra frame;
If the determination indicates a non-intra frame, encoding the first image as a non-intra frame;
Transmitting the encoded data;
The coding program that performs the processing.

Images