Moving Object Detection Method and Device

Abstract

The present disclosure provides a method and apparatus for detecting a moving object. The method for detecting a moving object includes: receiving an optical signal from a scene to be measured, and using the optical signal to perform optical imaging on the scene to be measured to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; within a predetermined time period, sequentially modulating the imaging optical signal by using modulation signals in a modulation signal set to generate a modulated optical signal, and generating a coded image based on the modulated optical signal; and using a moving object detection algorithm matching the modulation signal set to detect the coded image to identify the object to be detected in the coded image. The method for detecting a moving object provided by the present disclosure can identify the category of the moving object to be detected and multiple sets of position information in chronological order from a single coded image without the need for a high-speed camera, and can achieve efficient tracking and detection of high-speed moving objects.

Description

Technical Field

[0001] The present disclosure relates to the field of computer vision, and more particularly, to a moving object detection method and apparatus. Background Art

[0002] Object detection is one of the classic problems in the field of computer vision, and its task is usually to identify the position and category of objects in an image. With the development of machine learning technologies such as deep learning, object detection algorithms for static objects in static images have become increasingly mature. In practical applications, however, it is often necessary to track and detect moving objects. Traditionally, a video signal including a moving object can be generated, and object detection can be performed frame by frame on the video signal to achieve moving object detection. The prerequisite for accurately detecting a moving object is to ensure that there is no motion blur in each frame image of the video signal, which can usually be achieved by using a high-speed camera to capture the video. However, the use of a high-speed camera not only reduces the light flux of each frame image, but also significantly increases the data burden and cost of the system. Summary of the Invention

[0003] To solve the above problems, the present disclosure proposes a moving object detection algorithm and apparatus that can achieve the detection of moving objects without using a high-speed camera.

[0004] According to one aspect of the embodiments of the present disclosure, there is provided a moving object detection method, including: receiving an optical signal from a scene to be measured, and using the optical signal to perform optical imaging on the scene to be measured to generate an imaging optical signal of the scene to be measured, where the scene to be measured includes a moving object to be detected; within a predetermined time period, sequentially modulating the imaging optical signal by using modulation signals in a modulation signal set to generate a modulated optical signal, and generating an encoded image based on the modulated optical signal; and using a moving object detection algorithm matching the modulation signal set to detect the encoded image to identify the object to be detected in the encoded image.

[0005] According to an example of an embodiment of the present disclosure, wherein the modulation signal set includes a first number of modulation signals, the predetermined time period has a first duration, the duration of each modulation signal for modulating the imaging optical signal is a second duration, and the first duration is greater than or equal to the product of the first number and the second duration; wherein, modulating the imaging optical signal in sequence with the modulation signals in the modulation signal set within a predetermined time period to obtain a coded image includes: sequentially selecting each of the first number of modulation signals in the modulation signal set, and modulating the imaging optical signal with the modulation signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration, and sequentially obtaining a first number of modulated optical signals corresponding to the first number of modulation signals within the first duration; within the first duration, forming the coded image using the first number of modulated optical signals.

[0006] According to an example of an embodiment of the present disclosure, wherein modulating the imaging optical signal with the modulation signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration includes: inputting the modulation signal into a spatial light modulator, the spatial light modulator including a plurality of sub-units; adjusting the plurality of sub-units of the spatial light modulator with the modulation signal; and modulating the spatial distribution of the imaging optical signal with the adjusted plurality of sub-units to obtain a modulated optical signal corresponding to the modulation signal within the second duration.

[0007] According to an example of an embodiment of the present disclosure, wherein forming the coded image using the first number of modulated optical signals within the first duration includes: continuously collecting the first number of modulated optical signals with an image detector within the first duration, and generating the coded image based on the collected optical signals.

[0008] According to an example of an embodiment of the present disclosure, detecting the coded image using a moving object detection algorithm matching the modulation signal set to identify a to-be-detected object in the coded image includes: detecting the coded image to determine the position and category of the to-be-detected object in the coded image.

[0009] According to an example of an embodiment of the present disclosure, wherein determining the position and category of the to-be-detected object in the coded image includes: based on the coded image, determining a plurality of decoded images corresponding to each of the first number of modulation signals respectively; and determining the position and category of the to-be-detected object in the plurality of decoded images.

[0010] According to an example of an embodiment of the present disclosure, the set of modulation signals matching the moving object detection algorithm is determined through the following steps: obtaining a training data set, the training data set including a training image sequence and the labeled position information and labeled categories of one or more moving objects included in each training image in the training image sequence; using the labeled position information and labeled categories to perform supervised training on the moving object detection algorithm to determine the set of modulation signals.

[0011] According to an example of an embodiment of the present disclosure, the moving object detection algorithm includes a motion encoding module and a moving object detection module, and wherein, using the labeled position information and labeled categories to perform supervised training on the moving object detection algorithm to determine the set of modulation signals includes: using the motion encoding module to encode the training image sequence with a set of encoding signals to obtain training encoded images; using the moving object detection module to perform object detection on the training encoded images to obtain detection results; and using the labeled position information and the labeled categories to perform supervised training on the detection results to obtain a trained set of encoding signals, and determining the trained set of encoding signals as the set of modulation signals.

[0012] According to an example of an embodiment of the present disclosure, using the motion encoding module to encode the training image sequence with a set of encoding signals to obtain training encoded images includes: multiplying each encoding signal in the set of encoding signals with the corresponding training image in the training image sequence pixel by pixel, and summing the multiplication results to obtain the training encoded images.

[0013] According to another aspect of an embodiment of the present disclosure, there is provided a moving object detection device, including: an imaging lens configured to receive an optical signal from a scene to be measured and perform optical imaging on the scene to be measured with the optical signal to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; an encoding unit configured to, within a predetermined time period, modulate the imaging optical signal sequentially with modulation signals in a set of modulation signals to generate a modulated optical signal, and generate an encoded image based on the modulated optical signal; and a detection unit configured to detect the encoded image with a moving object detection algorithm matching the set of modulation signals to identify the object to be detected in the encoded image.

[0014] According to an example of an embodiment of the present disclosure, wherein the modulation signal set includes a first number of modulation signals, the predetermined time period has a first duration, the duration for each modulation signal to modulate the imaging optical signal is a second duration, and the first duration is greater than or equal to the product of the first number and the second duration. And wherein the modulation unit is further configured to sequentially select each of the first number of modulation signals in the modulation signal set, and use the modulation signal to modulate the imaging optical signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration, and sequentially obtain a first number of modulated optical signals corresponding to the first number of modulation signals within the first duration; and within the first duration, form the encoded image using the first number of modulated optical signals.

[0015] According to an example of an embodiment of the present disclosure, wherein the encoding unit includes a spatial light modulator, the spatial light modulator includes a plurality of sub-units, and is configured to: receive a modulation signal; adjust the plurality of sub-units of the spatial light modulator using the modulation signal; and modulate the spatial distribution of the imaging optical signal using the adjusted plurality of sub-units to obtain a modulated optical signal corresponding to the modulation signal within the second duration.

[0016] According to an example of an embodiment of the present disclosure, wherein the encoding unit further includes an image detector, the image detector is configured to: continuously collect the first number of modulated optical signals within the first duration, and generate the encoded image based on the collected optical signals.

[0017] According to an example of an embodiment of the present disclosure, wherein the detection unit is further configured to: detect the encoded image using a moving object detection algorithm matching the modulation signal set to determine the position and category of the object to be detected in the encoded image.

[0018] According to an example of an embodiment of the present disclosure, wherein the detection unit is further configured to: based on the encoded image, determine a plurality of decoded images corresponding to each of the first number of modulation signals respectively; and determine the position and category of the object to be detected in the plurality of decoded images.

[0019] According to an example of an embodiment of the present disclosure, wherein the modulation signal set matching the moving object detection algorithm is determined by the following steps: obtaining a training data set, the training data set including a training image sequence and the labeled position information and labeled category of one or more moving objects included in each training image in the training image sequence; using the labeled position information and labeled category to perform supervised training on the moving object detection algorithm to determine the modulation signal set.

[0020] According to an example of an embodiment of the present disclosure, wherein the moving object detection algorithm includes a motion encoding module and a moving object detection module, and wherein the moving object detection algorithm is supervised and trained using the labeled position information and labeled category to determine the modulation signal set, including: using the motion encoding module to encode the training image sequence with an encoding signal set to obtain a training encoded image; using the moving object detection module to perform object detection on the training encoded image to obtain a detection result; and using the labeled position information and labeled category to supervise and train the detection result to obtain a trained encoding signal set, and determining the trained encoding signal set as the modulation signal set.

[0021] According to an example of an embodiment of the present disclosure, wherein using the motion encoding module to encode the training image sequence with an encoding signal set to obtain a training encoded image includes: multiplying each encoding signal in the encoding signal set with the corresponding training image in the training image sequence pixel by pixel, and summing the multiplication results to obtain a training encoded image.

[0022] According to another aspect of an example of an embodiment of the present disclosure, there is provided a moving object detection device, including: an imaging lens configured to receive an optical signal from a scene to be measured and perform optical imaging on the scene to be measured using the optical signal to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; a spatial light modulator configured to receive a modulation signal set and modulate the imaging optical signal under the control of the modulation signal set to generate a modulated optical signal; an image detector configured to generate an encoded image based on the modulated optical signal; and one or more processors configured to: within a predetermined time period, sequentially provide the modulation signals in the modulation signal set to the spatial light modulator to control the spatial light modulator to modulate the imaging optical signal using the modulation signal to generate a modulated optical signal, and control the image detector to generate an encoded image based on the modulated optical signal; and use a moving object detection algorithm matching the modulation signal set to detect the encoded image to identify the object to be detected in the encoded image.

[0023] According to the moving object detection algorithms and devices in the above aspects of the embodiments of the present disclosure, by modulating the imaging optical signals of the scene to be measured in sequence using a modulation signal set to obtain a coded image, and using a moving object detection algorithm matching the modulation signal set to detect objects in the coded image, it is possible to identify the categories of the moving objects to be detected and multiple sets of position information in chronological order from a single coded image without the need for a high-speed camera, thereby realizing the tracking detection of high-speed moving objects, greatly improving the detection efficiency of moving objects, and reducing the system cost and data burden. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] By describing the embodiments of the present disclosure in more detail with reference to the accompanying drawings, the above and other objects, features, and advantages of the embodiments of the present disclosure will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure, and do not constitute a limitation to the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0025] Figure 1 Shows the overall architecture of a moving object detection system according to an example of an embodiment of the present disclosure;

[0026] Figure 2 Shows a flowchart of a moving object detection method according to an embodiment of the present disclosure;

[0027] Figure 3 Shows a schematic diagram of the training process of a moving object detection algorithm according to an example of an embodiment of the present disclosure;

[0028] Figure 4A Shows a schematic diagram of a detection result according to an example of an embodiment of the present disclosure;

[0029] Figure 4B Shows a schematic diagram of a detection result according to another example of an embodiment of the present disclosure;

[0030] Figure 4C Shows a schematic diagram of a detection result according to another example of an embodiment of the present disclosure;

[0031] Figure 5 Shows a schematic structural diagram of a moving object detection device according to an embodiment of the present disclosure; and

[0032] Figure 6 Shows a schematic structural diagram of a moving object detection device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] Next, the technical solutions in the embodiments of the present disclosure will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0034] In practical applications, it is often necessary to perform tracking and detection on moving objects. Especially for high-speed moving objects, it is often necessary to use a high-speed camera to capture videos for moving object detection to solve the motion blur problem caused by the high-speed movement of the object. The embodiments of the present disclosure provide a moving object detection method that can achieve the tracking and detection of high-speed moving objects without using a high-speed camera. In the moving object detection method according to the embodiments of the present disclosure, a single encoded image of a measurement scene containing a moving object can be obtained within any predetermined time period (for example, much longer than the exposure time of a high-speed camera), and the category of the moving object and multiple sets of position information in chronological order can be detected from the single encoded image, thereby realizing the efficient detection of moving objects, especially high-speed moving objects, greatly improving the detection efficiency of moving objects, and reducing the system cost and data burden.

[0035] The moving object detection method and device according to the embodiments of the present disclosure can be implemented, for example, as a moving object detection system including a hardware part and a software part. Figure 1 shows the overall architecture of an exemplary moving object detection system according to the embodiments of the present disclosure. The hardware part may include an imaging system 110 for generating an encoded image of a measurement scene containing an object to be detected, and the software part may include a moving object detection algorithm 120 for detecting the encoded image to identify the object to be detected. Figure 1 also shows an exemplary structure of the imaging system 110 and the moving object detection algorithm 120. As Figure 1 shown, the imaging system 110 may include, for example, an imaging lens, a spatial light modulator, an image detector, a relay lens, and other required devices, etc. The moving object detection algorithm 120 may include, for example, a moving object detection module 121, etc. However, the embodiments of the present disclosure are not limited thereto. The imaging system 110 and the moving object detection algorithm 120 may also include other required devices or structures. As Figure 1 shown, a modulation signal set matching the moving object detection algorithm 120 is input into the imaging system 110, and the imaging system 110 is used to generate an encoded image of the measurement scene. Then, the moving object detection algorithm 120 is used to detect the encoded image to obtain the detection result of the object to be detected.

[0036] Next, Figure 2 the moving object detection method according to the embodiments of the present disclosure will be specifically described.Figure 2 The flowchart of a moving object detection method 200 according to an embodiment of the present disclosure is shown.

[0037] As Figure 2 shown, in step S210, an optical signal from a scene to be measured is received, and the scene to be measured is optically imaged using the optical signal to generate an imaging optical signal of the scene to be measured. Among them, the scene to be measured includes a moving object to be detected, such as any type or number of moving objects such as a moving bicycle, a speeding car, a soaring plane, a running animal, etc. Here, the embodiments of the present disclosure do not make specific limitations on the type and number of the objects to be detected in the scene to be measured. According to an example of the embodiment of the present disclosure, an imaging lens can be used to receive the optical signal of the scene to be measured, and the received optical signal is used to optically image the scene to be measured to generate an imaging optical signal of the scene to be detected. The imaging lens can be a part of an imaging system, and for example, can be a convex lens, a concave lens, or various combinations thereof. The embodiments of the present disclosure do not make specific limitations on this. The generated imaging optical signal is a two-dimensional optical field signal representing the scene to be measured and varying with time. Then, the imaging optical signal can also be called a time-varying imaging optical signal, which can reflect the changes in the scene to be measured during imaging in real time, such as the movement process of the object to be detected in the scene to be measured.

[0038] In step S220, within a predetermined time period, the imaging optical signal is sequentially modulated using the modulation signals in the modulation signal set to generate a modulated optical signal, and an encoded image is obtained based on the modulated optical signal. Here, the predetermined time period has a first duration. The first duration can be, for example, the exposure time of the imaging system, and can be set arbitrarily according to actual needs. For example, the first duration can be set to be much longer than the exposure time of a high-speed camera, or can be set to other appropriate durations. The embodiments of the present disclosure do not make specific limitations on this.

[0039] According to an example of the embodiment of the present disclosure, the modulation signal set includes a first number of modulation signals, and each modulation signal can be a two-dimensional matrix corresponding to the imaging optical signal as a two-dimensional optical field signal. For example, a two-dimensional matrix composed of 0 and 1. Here, the first number can be set according to actual application needs. The embodiments of the present disclosure do not make specific limitations on this. According to an example of the embodiment of the present disclosure, the modulation signal set can be determined by performing machine learning training on a moving object detection algorithm. Then, the determined modulation signal set is matched with the moving object detection algorithm, which will be further described in detail below.

[0040] According to an example of an embodiment of the present disclosure, the modulation duration of each modulation signal in the modulation signal set for modulating the imaging optical signal is a second duration, and the first duration of the predetermined time period is greater than or equal to the product of the first quantity and the second duration. That is to say, within the predetermined time period, different modulation signals can be used to modulate the imaging optical signal multiple times. For example, when the first duration of the predetermined time period is equal to the product of the first quantity and the second duration, within the predetermined time period, the first quantity of modulation signals can be used to modulate the imaging optical signal multiple times, and the number of modulations is equal to the first quantity. Specifically, each modulation signal in the first quantity of modulation signals in the modulation signal set can be sequentially selected, and the selected modulation signal is used to modulate the imaging optical signal. As described above, the modulation duration of the imaging optical signal using each modulation signal is the second duration, so that the modulated optical signal corresponding to the selected modulation signal can be obtained within the second duration; multiple modulations are continuously performed within the predetermined time period, so that the first quantity of modulated optical signals corresponding to the first quantity of modulation signals can be obtained within the first duration of the predetermined time period. Subsequently, within the first duration, the first quantity of modulated optical signals obtained is used to form a coded image.

[0041] According to an example of an embodiment of the present disclosure, a spatial light modulator can be used to modulate the imaging optical signal using a modulation signal. A spatial light modulator is a device that modulates the spatial distribution of light waves. It can include a plurality of independent sub-units arranged in a one-dimensional array or a two-dimensional array. Each sub-unit can independently change its own optical properties, such as reflectivity, refractive index, transmittance, etc., under the control of an optical signal or an electrical signal, and then the spatial distribution of the light wave passing through them can be modulated using these sub-units. For example, a spatial light modulator can modulate optical parameters such as the amplitude, intensity, phase, and polarization state of light waves. The spatial light modulator can be a liquid crystal type spatial light modulator or a digital micromirror array, etc., and the embodiments of the present disclosure do not make specific limitations in this regard. In the embodiments of the present disclosure, each modulation signal in the modulation signal set can be used as a control signal for the spatial light modulator. Specifically, the modulation signal is input into the spatial light modulator, and the modulation signal is used to adjust the plurality of sub-units of the spatial light modulator. For example, when the modulation signal is a two-dimensional matrix, the values of each element in the modulation signal matrix can be used to respectively control the plurality of sub-units of the spatial light modulator to adjust the optical properties of the plurality of sub-units. Thus, the spatial distribution of the imaging optical signal can be modulated using the adjusted plurality of sub-units. For example, optical parameters such as the amplitude, intensity, phase, and polarization state of the imaging optical signal can be modulated to obtain a modulated optical signal corresponding to the input modulation signal within a modulation time of the second duration.

[0042] It should be noted that although a spatial light modulator is used to modulate the imaging optical signal in the above example, the embodiments of the present disclosure are not limited thereto, and any other device capable of changing the spatial distribution of the imaging optical signal may also be used.

[0043] As described above, after obtaining the first number of modulated optical signals corresponding to the first number of modulation signals within the first duration of the predetermined time period, an encoded image is formed using the obtained first number of modulated optical signals. According to an example of the embodiments of the present disclosure, an image detector may be used to continuously collect the first number of modulated optical signals within the first duration, and an encoded image may be generated based on the collected optical signals. Here, the image detector may be any device capable of converting an optical signal into an electrical signal, such as a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, etc., and the embodiments of the present disclosure do not make specific limitations thereto. For example, the image detector may continuously collect the modulated optical signals within the first duration through a relay lens (for example, it may be referred to as exposure), and generate an encoded image after performing photoelectric conversion and other processing on the collected optical signals. The relay lens may be, for example, a convex lens, a concave lens, or various combinations thereof, and the embodiments of the present disclosure do not make specific limitations thereto. Generally, the exposure time of the image detector is much longer than the modulation speed of the spatial light modulator. Therefore, in the case of using a spatial light modulator and an image detector, the exposure time of the imaging system according to the embodiments of the present disclosure will depend on the exposure time of the image detector. Therefore, the first duration of the predetermined time period may be set as the exposure time of the image detector.

[0044] After obtaining the encoded image, in step S230, a moving object detection algorithm matching the modulation signal set may be used to detect the encoded image to identify the object to be detected in the encoded image. According to an example of the embodiments of the present disclosure, the moving object detection algorithm may include a moving object detection module, and the moving object detection module may include, for example, a motion decoding module and an object detection module to perform motion decoding and object detection on the encoded image respectively. In the embodiments of the present disclosure, both the motion decoding module and the object detection module may be constructed using a neural network, and the object detection module may be implemented using an object recognition algorithm based on a neural network, such as a region convolutional neural network (RCNN), a faster region convolutional neural network (Faster-RCNN), a single shot multibox detector (SSD), etc., and the embodiments of the present disclosure do not make specific limitations thereto. As described above, the modulation signal set may be determined by performing machine learning training on the moving object detection algorithm, so the determined modulation signal set is matched with the moving object detection algorithm.

[0045] According to an example of an embodiment of the present disclosure, when detecting a coded image using a moving object detection algorithm, the category of the object to be detected can be identified. For example, when the object to be detected is a moving car, by detecting the coded image, the category of the object to be detected can be identified as "car". Additionally, using the moving object detection method according to an embodiment of the present disclosure, when detecting a coded image, the position of the object to be detected in the coded image can also be determined. After identifying the category and position of the object to be detected, the category of the object to be detected can be marked in the coded image. For example, a border can be used to select the identified object to be detected, and its category can be marked at the border (for example, marked as "car").

[0046] In addition, the moving object detection method according to an embodiment of the present disclosure can also detect multiple sets of position information and categories of the object to be detected from a single coded image. Specifically, when decoding a coded image using a moving object detection algorithm, multiple decoded images can be obtained based on a single coded image, and the multiple decoded images can respectively correspond to each of the modulation signals of the first number of modulation signals. According to an example of an embodiment of the present disclosure, the number of decoded images obtained from a single coded image can be equal to the first number of modulation signals. Object detection is respectively performed on the obtained multiple decoded images, and the position and category of the object to be detected in each of the multiple decoded images can be determined. Since the imaging optical signal of the scene to be measured changes over time, and each of the first number of modulation signals is sequentially selected within a predetermined time period to modulate the imaging optical signal to generate a coded image, the multiple decoded images respectively corresponding to the respective modulation signals correspond to the scene to be measured at different times. Therefore, the positions of the object to be detected in the multiple decoded images can reflect the movement trajectory of the object to be detected in the scene to be measured.

[0047] The following refers to Figure 3 to further describe a method for training a moving object detection algorithm to obtain a coded signal set. Figure 3 shows a schematic diagram of the training process of a moving object detection algorithm according to an example of an embodiment of the present disclosure. As Figure 3As shown, the moving object detection algorithm may include a moving object detection module 320, and the moving object detection module 320 may further include, for example, a motion decoding module 321 and an object detection module 322 to perform motion decoding and object detection on the encoded image respectively. The motion decoding module 321 and the object detection module 322 may be constructed using a neural network. For example, they may include network structures such as residual blocks and convolutional blocks. The embodiments of the present disclosure do not make specific limitations in this regard. In addition, the object detection module 322 may be implemented using an object recognition algorithm based on a neural network, such as RCNN, Faster-RCNN, SSD, etc. The embodiments of the present disclosure do not make specific limitations in this regard. Furthermore, the moving object detection algorithm may further include a motion encoding module 310. The motion encoding module 310 is a mathematical description of the hardware part of the moving object detection system including an imaging lens, a spatial light modulator, an image detector, etc. That is, the motion encoding module 310 may simulate the physical process of generating the encoded image. Therefore, a modulation signal set suitable for the moving object detection system can be obtained by performing machine learning training on the moving object detection algorithm including the motion encoding module 310.

[0048] When training the moving object detection algorithm, first, a training data set for training the moving object detection algorithm is obtained. The training data set may include a training image sequence, and each training image in the training image sequence contains annotation position information and annotation categories of one or more moving objects. Here, the training image sequence is a set of multiple training images that are continuous in time. For example, the training image sequence may be a video signal, and each frame image of the video signal serves as each training image in the training image sequence. The number of training images used for each training may be equal to the first number of the modulation signals described above, or may be greater than or equal to the first number. The embodiments of the present disclosure do not make specific limitations in this regard. For example, when the training image sequence included in the training data set has 80 training images and the first number is 8, 8 training images may be selected for each training. The training data set may be from a publicly available annotated data set for video object detection, such as the ImageNet VID (ImageNet Video Object Detection Data Set), etc. Then, the moving object detection algorithm is supervised and trained using the annotation position information and annotation categories in the training data set to determine the modulation signal set.

[0049] Specifically, as Figure 3As shown, first, the motion encoding module 310 is used to encode the training image sequence using the encoding signal set to obtain the training encoded image. As described above, the motion encoding module is a mathematical simulation of the physical process of generating the encoded image. In the example of step S220 above, the spatial light modulator sequentially selects modulation signal pairs in the modulation signal set within a predetermined time period to modulate the time-varying imaging optical signal, generating a first number of modulated optical signals. Then, the image detector continuously collects the first number of modulated optical signals to generate the encoded image. This step is equivalent to the process of multiplying and summing the time-varying imaging optical signal and the modulation signal set within a predetermined time period. Therefore, when training the moving object detection algorithm, the encoded signals in the encoding signal set can be multiplied pixel by pixel with the corresponding training images in the training image sequence, and the results of each multiplication are summed to obtain the training encoded image. Here, the encoding signal set used in the training process corresponds to the modulation signal set in the above step S220, that is, the trained encoding signal set can be used as the modulation signal set of the imaging system in the moving object detection system to modulate the imaging optical signal. As Figure 3 shown in the example of, the training image sequence including the moving hunting dog is multiplied and summed with the encoding signal set to obtain the training encoded image. It can be seen that after motion encoding, a series of training images including a clear image of the hunting dog are encoded into one image, in which the image of the hunting dog becomes blurred. Here, the encoded signals in the encoding signal set can be set to any suitable initial value, and the embodiments of the present disclosure do not make specific limitations on this.

[0050] Subsequently, the moving object detection module 320 is used to perform object detection on the training encoded image to obtain the detection result. Specifically, first, the motion decoding module 321 of the moving object detection module 320 can be used to decode the training encoded image to obtain a plurality of decoded images. The number of decoded images can, for example, correspond to the number of encoded signals in the encoding signal set and also correspond to the number of training images in the training image sequence. Then, the object detection module 322 of the moving object detection module 320 is used to perform object detection on the obtained plurality of decoded images to obtain the detection result. The detection result can, for example, be the category of one or more moving objects and multiple sets of position information in the plurality of decoded images, etc.

[0051] Since the class and position information of one or more moving objects included in each training image in the training image sequence are already labeled, the above detection results can be supervised and trained using the labeled class and labeled position information of the one or more moving objects. For example, the error between the labeled class and labeled position information and the detection results can be calculated, and the moving object detection algorithm can be supervised and trained by minimizing the error, thereby continuously optimizing the coding signal set and each network parameter in the moving object detection module until the optimal coding signal set and the optimal network parameters are obtained.

[0052] After obtaining the optimal coding signal set and the optimal network parameters through machine learning training of the moving object detection algorithm, the moving object detection algorithm is fixed. The trained coding signal set can be used as the modulation signal set and applied to the moving object detection system according to the embodiments of the present disclosure to image and modulate the scene to be detected containing the moving object to be detected to generate a coded image; then, the fixed moving object detection algorithm is used to detect the coded image to identify the object to be detected from the coded image. The specific steps are as described in the steps of the moving object detection method 200 referred to above Figure 2 and will not be elaborated here.

[0053] Figures 4A - 4C An example of the detection result obtained using the moving object detection method according to the embodiments of the present disclosure is shown. In Figure 4A the example, (a) is the coded image obtained using the moving object detection method according to the embodiments of the present disclosure, which includes a blurred image of a bicycle. The coded image is an image obtained by modulating and collecting the time-varying imaging optical signal of the scene to be detected containing the moving bicycle using eight modulation signals respectively. That is, the first number of the modulation signal set in the above step S220 is 8. After detecting the coded image using the moving object detection method according to the embodiments of the present disclosure, the class of the bicycle and the corresponding eight sets of position information can be identified from a single coded image. For the convenience of description, with the high-definition image sequence of the bicycle taken by a high-speed camera as the background, the detection results shown in (b)-(i) are shown, where the different positions of the bicycle are marked with rectangular frames and the class of the bicycle is marked in each image. In (b)-(i), the different positions and classes of the bicycle determined using the moving object detection method according to the embodiments of the present disclosure are marked in chronological order, that is, the movement trajectory of the bicycle is restored, and the tracking detection of the moving bicycle is realized.

[0054] In Figure 4BIn the example, (a) is an encoded image obtained by using the moving object detection method according to an embodiment of the present disclosure, which includes a blurred image of a car. The encoded image is an image obtained by modulating the time-varying imaging optical signal of the scene to be measured containing the moving car with eight encoding signals respectively and then performing image acquisition. That is to say, the first number of the modulation signal set in the above step S220 is 8. After detecting the encoded image by using the moving object detection method according to an embodiment of the present disclosure, the category of the car and eight corresponding sets of position information can be identified from a single encoded image, as shown in (b)-(i). Similarly, for the convenience of description, a high-definition image sequence of the car taken by a high-speed camera is used as the background of (b)-(i). In (b)-(i), the different positions and categories of the car determined by using the moving object detection method according to an embodiment of the present disclosure are marked in chronological order, that is, the motion trajectory of the moving car is restored, and the tracking detection of the moving car is realized.

[0055] In addition, the moving object detection method according to an embodiment of the present disclosure can realize the tracking detection of multiple moving objects simultaneously. In Figure 4C the example, (a) is an encoded image obtained by using the moving object detection method according to an embodiment of the present disclosure, which includes multiple blurred images of airplanes. The encoded image is an image obtained by modulating the time-varying imaging optical signal of the scene to be measured containing the moving airplanes with eight encoding signals respectively and then performing image acquisition. That is to say, the first number of the modulation signal set in the above step S220 is 8. After detecting the encoded image by using the moving object detection method according to an embodiment of the present disclosure, the category of each airplane and eight corresponding sets of position information of each airplane can be identified from a single encoded image, as shown in (b)-(i). Similarly, for the convenience of description, a high-definition image sequence of the airplanes taken by a high-speed camera is used as the background of (b)-(i). In (b)-(i), the different positions and categories of each airplane determined by using the moving object detection method according to an embodiment of the present disclosure are marked in chronological order, that is, the motion trajectories of multiple airplanes are restored simultaneously, and the tracking detection of the high-speed moving airplanes is realized.

[0056] From the above description and Figures 4A - 4CAs can be seen from the examples, when using the moving object detection method according to an embodiment of the present disclosure, a single encoded image of the moving object can be generated within a relatively long exposure time, and the category of the moving object and multiple sets of position information in chronological order can be detected from the single encoded image, thereby greatly improving the efficiency of object detection. Especially when tracking and detecting a high-speed moving object, using the moving object detection method according to an embodiment of the present disclosure can restore the motion trajectory of the high-speed moving object over a long period of time with only a few encoded images captured, and there is no need to use a high-speed camera, which significantly reduces the system cost and data burden while achieving efficient detection of high-speed moving objects.

[0057] Next, refer to Figure 5 Describe the moving object detection device according to an embodiment of the present disclosure. Figure 5 FIG. shows a schematic structural diagram of a moving object detection device 500 according to an embodiment of the present disclosure. Since the moving object detection device 500 is the same as the details of the moving object detection method 200 described above in conjunction with Figure 2 For simplicity, the detailed description of the same content is omitted here. As Figure 5 shown, the moving object detection device 500 includes an imaging lens 510, an encoding unit 520, and a detection unit 530. In addition to these three units, the device 500 may also include other components. However, since these components are not relevant to the content of the embodiments of the present disclosure, their illustrations and descriptions are omitted here.

[0058] The imaging lens 510 is configured to receive an optical signal from a scene to be measured and perform optical imaging on the scene to be measured using the optical signal to generate an imaging optical signal of the scene to be measured. Among them, the scene to be measured includes a moving object to be detected, such as any type or number of moving objects such as a moving bicycle, a speeding car, a soaring plane, a running animal, etc. Here, the embodiments of the present disclosure do not specifically limit the type and number of the objects to be detected. The imaging lens may be a part of an imaging system and may be, for example, a convex lens, a concave lens, or various combinations thereof. The embodiments of the present disclosure do not specifically limit this. The generated imaging optical signal is a two-dimensional optical field signal representing the scene to be measured and changing with time, and the imaging optical signal may also be referred to as a time-varying imaging optical signal, which can reflect the changes in the scene to be measured during imaging in real time, such as the motion process of the object to be detected in the scene to be measured.

[0059] The encoding unit 520 is configured to modulate the imaging optical signal sequentially using the modulation signals in the modulation signal set within a predetermined time period to obtain an encoded image. Here, the predetermined time period has a first duration, and the first duration can be, for example, the exposure time of the imaging system and can be set arbitrarily according to actual requirements. For example, the first duration can be set to be much greater than the exposure time of the high-speed camera, or can be set to other appropriate durations. The embodiments of the present disclosure do not make specific limitations on this.

[0060] According to an example of the embodiments of the present disclosure, the modulation signal set includes a first number of modulation signals, where each modulation signal can be a two-dimensional matrix corresponding to the imaging optical signal as a two-dimensional light field signal. For example, a two-dimensional matrix composed of 0 and 1. Here, the first number can be set according to actual application requirements, and the embodiments of the present disclosure do not make specific limitations on this. According to an example of the embodiments of the present disclosure, the modulation signal set can be determined by performing machine learning training on the moving object detection algorithm, and the determined modulation signal set is matched with the moving object detection algorithm. The details of performing machine learning training on the moving object detection algorithm to determine the modulation signal set are similar to the process described above with reference to Figure 3 The description is similar, so the repeated description of the same content is omitted here.

[0061] According to an example of the embodiments of the present disclosure, the duration for each modulation signal in the modulation signal set to modulate the imaging optical signal is a second duration, and the first duration of the predetermined time period is greater than or equal to the product of the first number and the second duration. That is, within the predetermined time period, the encoding unit 520 can modulate the imaging optical signal multiple times using different modulation signals respectively. For example, when the first duration of the predetermined time period is equal to the product of the first number and the second duration, within the predetermined time period, the encoding unit 520 can modulate the imaging optical signal multiple times using the first number of modulation signals, and the number of modulations is equal to the first number. Specifically, the encoding unit 520 can sequentially select each of the first number of modulation signals in the modulation signal set and modulate the imaging optical signal using the selected modulation signal. As described above, the duration for modulating the imaging optical signal using each modulation signal is the second duration, so the modulated optical signal corresponding to the selected modulation signal can be obtained within the second duration; the encoding unit 520 continuously performs multiple modulations within the predetermined time period, so that within the first duration of the predetermined time period, the first number of modulated optical signals corresponding to the first number of modulation signals are obtained respectively. Subsequently, within the first duration, the first number of modulated optical signals obtained are used to form an encoded image.

[0062] According to an example of an embodiment of the present disclosure, the encoding unit 520 may include, for example, a spatial light modulator. A spatial light modulator is a device that modulates the spatial distribution of light waves. It may include a plurality of independent sub-units arranged in a one-dimensional array or a two-dimensional array. Each sub-unit can independently change its own optical properties, such as reflectivity, refractive index, transmittance, etc., under the control of an optical signal or an electrical signal. Furthermore, these sub-units can be used to modulate the spatial distribution of the light waves passing through them. For example, a spatial light modulator can modulate optical parameters such as the amplitude, intensity, phase, and polarization state of light waves. The spatial light modulator can be a liquid crystal type spatial light modulator or a digital micromirror array, etc., and the embodiments of the present disclosure do not make specific limitations thereto. In the embodiments of the present disclosure, each modulation signal in the modulation signal set can be used as a control signal for the spatial light modulator. Specifically, the spatial light modulator is configured to receive the modulation signal and use the modulation signal to adjust the plurality of sub-units of the spatial light modulator. For example, in the case where the modulation signal is a two-dimensional matrix, the values of the respective elements in the modulation signal matrix can be used to separately control the plurality of sub-units of the spatial light modulator to adjust the optical properties of the plurality of sub-units. Thus, the spatial light modulator can use the adjusted plurality of sub-units to modulate the spatial distribution of the imaging light signal. For example, it can modulate optical parameters such as the amplitude, intensity, phase, and polarization state of the imaging light signal to obtain a modulated light signal corresponding to the input modulation signal within a modulation time of a second duration.

[0063] It should be noted that although a spatial light modulator is used to modulate the imaging light signal in the above example, the embodiments of the present disclosure are not limited thereto. The encoding unit 520 may also include any other device capable of changing the spatial distribution of the imaging light signal.

[0064] According to an example of an embodiment of the present disclosure, the encoding unit 520 may further include an image detector configured to continuously collect a first number of modulated optical signals within a first time period and generate an encoded image based on the collected optical signals. Here, the image detector may be any device capable of converting an optical signal into an electrical signal, such as a charge-coupled device (CCD) sensor, a complementary metal oxide semiconductor (CMOS) sensor, etc., and the embodiments of the present disclosure do not make specific limitations in this regard. For example, the image detector may continuously collect the modulated optical signals within the first time period through a relay lens (for example, it can be called exposure), and generate an encoded image after performing photoelectric conversion and other processing on the collected optical signals. The relay lens may be, for example, a convex lens, a concave lens, or various combinations thereof, and the embodiments of the present disclosure do not make specific limitations in this regard. Generally, the exposure time of the image detector is much longer than the modulation speed of the spatial light modulator. Therefore, in the case of using a spatial light modulator and an image detector, the exposure time of the imaging system according to the embodiments of the present disclosure will depend on the exposure time of the image detector. Therefore, the first time period of the predetermined time period may be equal to the exposure time of the image detector.

[0065] The detection unit 530 is configured to detect the encoded image by using a moving object detection algorithm that matches the modulation signal set to identify the object to be detected in the encoded image. According to an example of an embodiment of the present disclosure, the moving object detection algorithm may include a moving object detection module. Among them, the moving object detection module may include, for example, a motion decoding module and an object detection module to perform motion decoding and object detection on the encoded image respectively. Both the motion decoding module and the object detection module can be constructed by a neural network, and the object detection module can be implemented by using an object recognition algorithm based on a neural network, such as RCNN, Faster-RCNN, SSD, etc., and the embodiments of the present disclosure do not make specific limitations in this regard. As described above, the modulation signal set can be determined by machine learning training of the moving object detection algorithm, so the determined modulation signal set matches the moving object detection algorithm.

[0066] According to an example of an embodiment of the present disclosure, when the detection unit 530 detects the encoded image by using the moving object detection algorithm, it can identify the category of the object to be detected. For example, when the object to be detected is a moving car, by detecting the encoded image, the category of the object to be detected can be identified as "car". In addition, by using the moving object detection device according to the embodiments of the present disclosure, when the detection unit 530 detects the encoded image, it can also determine the position of the object to be detected in the encoded image. After identifying the category and position of the object to be detected, the category of the object to be detected can be marked in the encoded image. For example, a border can be used to select the identified object to be detected, and its category can be marked at the border (for example, marked as "car").

[0067] In addition, the moving object detection device according to an embodiment of the present disclosure can detect multiple sets of position information of an object to be detected from a single encoded image. Specifically, when the detection unit 530 decodes the encoded image using a moving object detection algorithm, multiple decoded images can be obtained based on a single encoded image, and the multiple decoded images can respectively correspond to each of the modulation signals in the first number of modulation signals. According to an example of the embodiment of the present disclosure, the number of decoded images obtained from a single encoded image can be equal to the first number of modulation signals. The detection unit 530 respectively performs object detection on the obtained multiple decoded images, and can determine the position and category of the object to be detected in the multiple decoded images. Since the imaging optical signal of the scene to be measured changes with time, and each of the modulation signals in the first number of modulation signals is sequentially selected within a predetermined time period to modulate the imaging optical signal to generate an encoded image, the multiple decoded images respectively corresponding to the modulation signals correspond to the scene to be measured at different times. Therefore, the positions of the object to be detected in the multiple decoded images can reflect the movement trajectory of the object to be detected, as Figures 4A - 4C shown in the example detection result in

[0068] By using the moving object detection device according to the above embodiment of the present disclosure, a single encoded image of a moving object can be generated within a relatively long exposure time, and the category of the moving object and multiple sets of position information in chronological order can be detected from a single encoded image, thereby greatly improving the efficiency of object detection. Especially when performing tracking detection on a high-speed moving object, using the moving object detection device according to the embodiment of the present disclosure can recover the movement trajectory of the high-speed moving object over a long time with only a few encoded images captured, and there is no need to use a high-speed camera, which greatly reduces the system cost and data burden while achieving efficient detection of high-speed moving objects.

[0069] Next, Figure 6 the moving object detection device according to an embodiment of the present disclosure will be described with reference to Figure 6 FIG. shows a schematic structural diagram of a moving object detection device 600 according to an embodiment of the present disclosure. Since the moving object detection device 600 has the same details as the moving object detection method 200 described above in conjunction with Figure 2 , for the sake of simplicity, the detailed description of the same content is omitted here. As Figure 6 shown, the moving object detection device 600 can include an imaging lens 610, a spatial light modulator 620, an image detector 630, and one or more processors 640. In addition to these four units, the moving object detection device 600 can also include other components, for example, one or more storage devices, input / output components, etc., and the embodiments of the present disclosure do not make specific limitations on this.

[0070] The imaging lens 610 is configured to receive an optical signal from a scene to be measured, and perform optical imaging on the scene to be measured by using the optical signal to generate an imaging optical signal of the scene to be measured. Here, the step of generating the imaging optical signal is similar to the step S210 of the moving object detection method described above with reference to Figure 2 and the details of the function of the imaging lens 510 described with reference to Figure 5 Therefore, for simplicity, the repeated description of the same content is omitted here. The spatial light modulator 620 is configured to receive a set of modulation signals, and modulate the imaging optical signal under the control of the set of modulation signals to generate a modulated optical signal; the image detector 630 is configured to generate a coded image based on the modulated optical signal. Here, the step of generating the coded image is similar to the step S220 of the moving object detection method described above with reference to Figure 2 and the details of the function of the coding unit 520 described with reference to Figure 5 Therefore, for simplicity, the repeated description of the same content is omitted here.

[0071] One or more processors 640 are configured to sequentially provide the modulation signals in the set of modulation signals to the spatial light modulator 620 within a predetermined time period, so as to control the spatial light modulator 620 to modulate the imaging optical signal by using the modulation signals to generate a modulated optical signal, and control the image detector 630 to generate a coded image based on the modulated optical signal. In addition, one or more processors 640 are further configured to detect the coded image by using a moving object detection algorithm matching the set of coded signals, so as to identify an object to be detected in the coded image. Here, the step of detecting the coded image is similar to the step S230 of the moving object detection method described above with reference to Figure 2 and the details of the function of the detection unit 530 described with reference to Figure 5 Therefore, for simplicity, the repeated description of the same content is omitted here.

[0072] Those skilled in the art can understand that the content disclosed in the present disclosure can have various variations and improvements. For example, the various devices or components described above can be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0073] In addition, as shown in this disclosure and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include the plural. The words "first", "second", and similar terms used in this disclosure do not denote any order, quantity, or importance, but are merely used to distinguish different components. Similarly, words such as "comprising" or "including" mean that the elements or items appearing before the word cover the elements or items listed after the word and their equivalents, without excluding other elements or items. The words "connected" or "coupled" and the like are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0074] In addition, flowcharts are used in this disclosure to illustrate the operations performed by the systems of the embodiments according to the embodiments of this disclosure. It should be understood that the operations before or below are not necessarily executed precisely in order. Instead, the various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be superimposed on these processes, or one or more steps can be removed from these processes.

[0075] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. It should also be understood that terms such as those defined in a general dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0076] The above has described this disclosure in detail, but for those skilled in the art, it is obvious that this disclosure is not limited to the embodiments described in this specification. This disclosure can be implemented in the form of modifications and changes without departing from the spirit and scope of this disclosure as determined by the claims. Therefore, the description in this specification is for the purpose of illustration and has no limiting significance for this disclosure.

Claims

1. A method for detecting a moving object, comprising: receiving an optical signal from a scene to be measured, and performing optical imaging on the scene to be measured by using the optical signal to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; within a predetermined time period, sequentially modulating the imaging optical signal by using modulation signals in a modulation signal set to generate a modulated optical signal, continuously collecting the modulated optical signal by using an image detector within the predetermined time period, and generating a coded image based on the collected optical signal; and detecting the coded image by using a moving object detection algorithm matching the modulation signal set to identify the object to be detected in the coded image, wherein the modulation signal set matching the moving object detection algorithm is determined by the following steps: acquiring a training data set, the training data set including a training image sequence and labeled position information and labeled categories of one or more moving objects included in each training image in the training image sequence; supervising and training the moving object detection algorithm by using the labeled position information and labeled categories to determine the modulation signal set.

2. The method for detecting a moving object according to claim 1, wherein the modulation signal set includes a first number of modulation signals, the predetermined time period has a first duration, the duration of modulating the imaging optical signal by each modulation signal is a second duration, and the first duration is greater than or equal to the product of the first number and the second duration; wherein sequentially modulating the imaging optical signal by using modulation signals in the modulation signal set within a predetermined time period to generate a modulated optical signal includes: sequentially selecting each of the first number of modulation signals in the modulation signal set, and modulating the imaging optical signal by using the modulation signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration, and sequentially obtaining a first number of modulated optical signals respectively corresponding to the first number of modulation signals within the first duration.

3. The method for detecting a moving object according to claim 2, wherein modulating the imaging optical signal by using the modulation signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration includes: inputting the modulation signal into a spatial light modulator, the spatial light modulator including a plurality of sub-units; adjusting the plurality of sub-units of the spatial light modulator by using the modulation signal; and modulating the spatial distribution of the imaging optical signal by using the adjusted plurality of sub-units to obtain a modulated optical signal corresponding to the modulation signal within the second duration.

4. The method for detecting a moving object according to claim 1, detecting the coded image by using a moving object detection algorithm matching the modulation signal set to identify the object to be detected in the coded image comprising: detecting the coded image to determine the position and category of the object to be detected in the coded image.

5. The method for detecting a moving object according to claim 4, wherein Determining the position and category of the object to be detected in the encoded image includes: Based on the encoded image, determining a plurality of decoded images corresponding to each modulation signal in the modulation signal set; and Determining the position and category of the object to be detected in the plurality of decoded images.

6. The moving object detection method according to claim 1, wherein, the moving object detection algorithm includes a motion encoding module and a moving object detection module, and wherein, using the labeled position information and labeled category to perform supervised training on the moving object detection algorithm to determine the modulation signal set includes: Using the motion encoding module to encode the training image sequence with an encoding signal set to obtain a training encoded image; Using the moving object detection module to perform object detection on the training encoded image to obtain a detection result; and Using the labeled position information and the labeled category to perform supervised training on the detection result to obtain a trained encoding signal set, and determining the trained encoding signal set as the modulation signal set.

7. The moving object detection method according to claim 6, wherein, Using the motion encoding module to encode the training image sequence with an encoding signal set to obtain a training encoded image includes: Multiplying each encoding signal in the encoding signal set with the corresponding training image in the training image sequence pixel by pixel, and summing the multiplication results to obtain the training encoded image.

8. A moving object detection device, comprising: An imaging lens configured to receive an optical signal from a scene to be measured, and perform optical imaging on the scene to be measured with the optical signal to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; An encoding unit configured to, within a predetermined time period, modulate the imaging optical signal sequentially with modulation signals in a modulation signal set to generate a modulated optical signal, continuously collect the modulated optical signal with an image detector within the predetermined time period, and generate an encoded image based on the collected optical signal; and A detection unit configured to use a moving object detection algorithm matching the modulation signal set to detect the encoded image to identify the object to be detected in the encoded image, wherein the modulation signal set matching the moving object detection algorithm is determined by the following steps: Obtaining a training data set, the training data set including a training image sequence and labeled position information and labeled category of one or more moving objects included in each training image in the training image sequence; Using the labeled position information and labeled category to perform supervised training on the moving object detection algorithm to determine the modulation signal set.

9. The moving object detection device according to claim 8, wherein, The modulation signal set includes a first number of modulation signals, the predetermined time period has a first duration, the duration for each modulation signal to modulate the imaging optical signal is a second duration, and the first duration is greater than or equal to the product of the first number and the second duration. And wherein, the moving object detection device further includes a modulation unit, and the modulation unit is configured to, sequentially select each of the first number of modulation signals in the modulation signal set, and use the modulation signal to modulate the imaging optical signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration, and sequentially obtain a first number of modulated optical signals corresponding to the first number of modulation signals within the first duration.

10. The moving object detection device according to claim 9, wherein, the encoding unit includes a spatial light modulator, the spatial light modulator includes a plurality of sub-units, and is configured to: receive a modulation signal; use the modulation signal to adjust the plurality of sub-units of the spatial light modulator; and use the adjusted plurality of sub-units to modulate the spatial distribution of the imaging optical signal to obtain a modulated optical signal corresponding to the modulation signal within the second duration.

11. The moving object detection device according to claim 8, wherein, the detection unit is further configured to: use a moving object detection algorithm matching the modulation signal set to detect the encoded image to determine the position and category of the object to be detected in the encoded image.

12. The moving object detection device according to claim 11, wherein, the detection unit is further configured to: determine a plurality of decoded images respectively corresponding to the respective modulation signals in the modulation signal set based on the encoded image; and determine the position and category of the object to be detected in the plurality of decoded images.

13. The moving object detection device according to claim 8, wherein, the moving object detection algorithm includes a motion encoding module and a moving object detection module, and wherein, the modulation signal set is determined by supervising and training the moving object detection algorithm using the labeled position information and labeled category, including: using the motion encoding module to encode the training image sequence using an encoding signal set to obtain a training encoded image; using the moving object detection module to perform object detection on the training encoded image to obtain a detection result; and using the labeled position information and labeled category to perform supervising and training on the detection result to obtain a trained encoding signal set, and determining the trained encoding signal set as the modulation signal set.

14. The moving object detection device according to claim 13, wherein, using the motion encoding module to encode the training image sequence using an encoding signal set to obtain a training encoded image includes: multiplying each encoding signal in the encoding signal set with the corresponding training image in the training image sequence pixel by pixel, and summing the multiplication results to obtain a training encoded image.

15. A moving object detection device, comprising: An imaging lens configured to receive an optical signal from a scene to be measured and optically image the scene to be measured using the optical signal to generate an imaging optical signal of the scene to be measured, wherein the scene to be measured includes a moving object to be detected; A spatial light modulator configured to receive a set of modulation signals and modulate the imaging optical signal under the control of the set of modulation signals to generate a modulated optical signal; An image detector configured to generate a coded image based on the modulated optical signal; And One or more processors configured to: Within a predetermined time period, sequentially provide the modulation signals in the set of modulation signals to the spatial light modulator to control the spatial light modulator to modulate the imaging optical signal using the modulation signals to generate a modulated optical signal, and control the image detector to continuously collect the modulated optical signal within the predetermined time period and generate a coded image based on the collected optical signal; Detect the coded image using a moving object detection algorithm matching the set of modulation signals to identify the object to be detected in the coded image, wherein the set of modulation signals matching the moving object detection algorithm is determined by the following steps: Obtain a training data set, the training data set including a training image sequence and labeled position information and labeled categories of one or more moving objects included in each training image in the training image sequence; Supervisedly train the moving object detection algorithm using the labeled position information and labeled categories to determine the set of modulation signals.

Images