Portable multispectral video image recording and analysis processing system and method

Abstract

The application belongs to the field of video image processing of target detection and identification, and particularly relates to a portable multispectral video image recording and analysis processing system and method. The system comprises a continuous zoom optical system, an image analysis processing system connected to the continuous zoom optical system, and an imaging control system. The continuous zoom optical system comprises multiple imaging cameras, which are used to receive parameter information controlled by the imaging control system, adjust the focal length of the optical lens of the camera, and send images of different wave bands to the image analysis processing system for processing. After the image analysis processing system processes the images, parameter information of the continuous zoom optical system is obtained and sent to the imaging control system, and the processed images are output to a terminal in a set format. According to the result information, the analysis control module analyzes the parameter information of each imaging camera in the continuous zoom optical system, and controls the lens according to the analysis of the target in the image.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of video image processing of target detection and identification, and particularly relates to a portable multi-spectral video image recording and analysis processing system and method. BACKGROUND

[0002] Target detection and identification is the basis of image and video processing, and is widely applied in various fields such as military, medicine, industry, commerce and education. The most important point in target detection and identification is the acquisition of target feature information. In the selection of video acquisition and processing systems in the early data collection and analysis and the later system of target detection, identification and tracking projects, it is always desired to collect the features of the target in various environments that are more distinct from the background. On the one hand, for some targets, it is impossible to use the prior knowledge of various targets to distinguish the class attributes of unknown targets, and the features are significant when imaging in that waveband. On the other hand, the imaging information based on a single waveband cannot significantly reflect the difference from the background, and is easily affected by other external factors. The inherent differences and complementarity of multi-waveband imaging information can effectively improve the performance of target detection and identification. The current video acquisition and processing methods are basically based on specific targets and specific tasks, and do not have universality. Therefore, how to realize multi-waveband simultaneous imaging of targets in different sizes, distances, motion conditions and different environments, convenient and efficient target data recording and analysis, shorten the project research and verification period, and greatly reduce the cost, is a technical problem to be solved in the current technology. SUMMARY

[0003] The present application aims to provide a portable multi-spectral video image recording and analysis processing system and method to overcome the defect that the imaging information of a single waveband in the above-mentioned existing target identification method cannot significantly reflect the difference from the background, and is easily affected by other external factors.

[0004] The technical solution adopted by the present application to achieve the above-mentioned purpose is as follows: a portable multi-spectral video image recording and analysis processing system, comprising: a continuous zoom optical system and an image analysis processing system and an imaging control system connected thereto.

[0005] The continuous zoom optical system comprises a plurality of imaging cameras, which are used to receive the parameter information controlled by the imaging control system, and provide the optical lens focal length of different imaging cameras required by the target at different distances, simultaneously collect images of different wavebands and send them to the image analysis processing system for processing.

[0006] The image analysis processing system is used to receive and analyze the images of different wavebands collected by the continuous zoom optical system, acquire the parameter information of the continuous zoom optical system, and send it to the imaging control system, and at the same time, output it to the terminal in a set format.

[0007] The imaging control system is configured to receive parameter information of the continuous zoom optical system of the image analysis processing system, and control optical lens focal lengths of different imaging cameras of the continuous zoom optical system; and set image acquisition time points of different wavebands for the continuous zoom optical system, so as to ensure that ending time points of imaging of the continuous zoom optical system are the same.

[0008] The continuous zoom optical system comprises a plurality of zoom lens control systems and a plurality of imaging cameras.

[0009] The imaging cameras comprise visible light cameras, short-wave cameras, medium-wave cameras and long-wave cameras.

[0010] The corresponding visible light cameras, short-wave cameras, medium-wave cameras and long-wave cameras are connected to the corresponding zoom lens control systems to adjust lens focal lengths of the corresponding cameras.

[0011] Each of the zoom lens control systems is connected to the imaging control system, receives parameter information of the imaging cameras sent by the imaging control system to adjust the lens focal lengths of the imaging cameras, and receives the set image acquisition time points of different wavebands to ensure that the ending time points of imaging of the continuous zoom optical system are the same.

[0012] The image analysis processing system comprises an image preprocessing module, an image stitching module, a target detection and recognition module, an analysis control module and a video output module.

[0013] The image preprocessing module is configured to pre-process different waveband images received by the continuous zoom optical system, and send the pre-processed images of different wavebands to the image stitching module and the target detection and recognition module.

[0014] The image stitching module is configured to stitch the images of different wavebands into one picture and form a video stream, and output the video stream to a terminal in a set video format.

[0015] The target detection and recognition module is configured to perform target recognition on the pre-processed images of different wavebands, calculate result information of the target, and send the pre-processed images and the result information to the analysis control module.

[0016] The analysis control module is configured to analyze the result information to obtain parameter information of the imaging cameras in the continuous zoom optical system, and return the parameter information to the imaging control system for real-time control.

[0017] The pre-processed images of the same waveband are tracked by using a gate tracking method or a feature matching method, and image streams in a tracking process are sent to the video output module.

[0018] The images preprocessed in different wave bands are fused by using an HSV fusion algorithm or a contour fusion algorithm, and the image stream obtained in the fusion process is sent to a video output module;

[0019] The video output module is configured to output in a set format according to the selection of the analysis control module.

[0020] The result information includes target size, target motion direction, and target motion rate.

[0021] The parameter information includes zoom lens focal length size, focal length step and frequency, and imaging camera shutter value, gain setting value, and adjustment direction.

[0022] The imaging control system includes an imaging control module and an image synchronization control module connected to each imaging camera of the continuous zoom optical system, respectively.

[0023] The imaging control module is configured to receive the parameter information of each imaging camera in the continuous zoom optical system sent by the analysis control module, and control according to the parameter information of each imaging camera.

[0024] The image synchronization control module is configured to set the image video acquisition time points of each imaging camera in the continuous zoom optical system, so as to ensure that the ending time points of imaging are the same.

[0025] A processing method of a portable multi-spectral video image recording and analysis processing system includes the following steps:

[0026] 1) The continuous zoom optical system collects images of different wave bands of a target and sends them to an image preprocessing module.

[0027] 2) The image preprocessing module pre-processes the images of different wave bands and sends them to an image stitching module and a target detection and recognition module, respectively.

[0028] 3) The image stitching module stitches the images of different wave bands at the same time into one picture by using a multi-wave band stitching method, forms a video stream, and outputs the video stream to a video stream storage terminal at a set unified resolution.

[0029] Meanwhile, the target detection and recognition module performs target recognition on the images preprocessed in different wave bands, calculates result information of the target, and sends the result information and the extracted target features after preprocessing to an analysis control module.

[0030] 4) The analysis control module analyzes the result information and obtains parameter information of each imaging camera in the continuous zoom optical system, which is fed back to an imaging control system to perform real-time control on each imaging camera in the continuous zoom optical system.

[0031] Meanwhile, the analysis control module uses a gate tracking method or a feature matching method to track the preprocessed images of the same waveband, and sends the image stream obtained in the tracking process to the video output module;

[0032] The analysis control module uses an HSV fusion algorithm or a contour fusion algorithm to fuse the preprocessed images of different wavebands, and sends the image stream obtained in the fusion process to the video output module;

[0033] 5) The video output module outputs in a set format according to the selection of the analysis control module.

[0034] The image preprocessing module pre-processes the images of different wavebands, specifically as follows:

[0035] The images collected by each imaging camera are sequentially subjected to denoising, filtering, image enhancement, edge extraction, and fog removal processing, to obtain multiple preprocessed images of different wavebands.

[0036] In step 3), the target detection and recognition module detects and recognizes the targets in the preprocessed images of different wavebands, specifically as follows:

[0037] (1) The image preprocessing module obtains secondary images of different wavebands, and applies a foreground and background segmentation method to the secondary images, and the target detection and recognition module obtains target features such as color, grayscale value, gradient, and histogram, and simultaneously segments out background information.

[0038] (2) After detecting and recognizing the target, the target size, motion direction, and motion rate of the target are calculated.

[0039] The method for obtaining the target size, motion direction, and motion rate according to the target features is as follows:

[0040] The background segmentation method is used to extract the target features. First, a background model is established, and the features of the image sequence are compared to segment out the background and the foreground, thereby obtaining the target region and edges, and further obtaining the target size.

[0041] Meanwhile, the target features such as color, grayscale value, gradient, and histogram are obtained. The position of the target in the image in multiple frames can be used to determine the motion direction of the target, and the motion rate of the target can be obtained by combining the time of each frame.

[0042] In step 4), the analysis control module analyzes the result information to obtain the parameter information of each imaging camera in the continuous zoom optical system, specifically as follows:

[0043] According to the target size, the size of the zoom lens focus is obtained, and the target size is ensured to be kept in the set range at the focus, that is, the focus is increased when the target is reduced, and the focus is reduced when the target is increased;

[0044] According to the target motion direction, the target motion rate and the target size, the step and the frequency of changing the focus are determined, that is, the larger the target or the larger the target motion rate, the larger the step and the higher the frequency of changing the focus, and vice versa, the smaller the target or the smaller the motion rate, the smaller the step and the lower the frequency of changing the focus;

[0045] According to the brightness of the target, the average gray value of the background and the histogram, the shutter value, the gain setting value and the adjustment direction of the corresponding imaging camera are determined.

[0046] The present application has the following beneficial effects and advantages:

[0047] 1. The present application introduces a multi-spectral optical imaging system, and uses a continuous zoom lens to meet different image information in multiple spectral bands under different distances and different sizes of targets under different imaging parameters.

[0048] 2. In the present application, the image analysis processing system uses detection, recognition and tracking on the target in the image processing stage, selects the analysis control module according to the video output module, and outputs in the set format, so as to verify the existing algorithm synchronously, so that the target data recording and analysis are convenient and fast, the pre-research verification period is shortened, and the cost is greatly reduced. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The structural schematic diagram of the present application;

[0050] Figure 2 The system block diagram and method flow schematic diagram of the present application;

[0051] Figure 3 The image splicing module splicing schematic diagram of the present application. DETAILED DESCRIPTION

[0052] The present application will be further described in detail below in combination with the drawings and embodiments.

[0053] As Figure 2 shown, the system block diagram and method flow schematic diagram of the present application, wherein the present application comprises a continuous zoom optical system and an image analysis processing system and an imaging control system connected therewith;

[0054] The continuous zoom optical system comprises a plurality of imaging cameras, which are used for receiving the parameter information controlled by the imaging control system, and providing the optical lens focus of different imaging cameras required by the target at different distances, and collecting images of different wave bands and sending them to the image analysis processing system for processing;

[0055] An image analysis processing system is configured to receive images of different wave bands captured by a continuous zoom optical system, analyze and process the images, obtain parameter information of the continuous zoom optical system, and send the parameter information to an imaging control system, and output the images to a terminal in a set format.

[0056] An imaging control system is configured to receive the parameter information of the continuous zoom optical system from the image analysis processing system, and control focal lengths of optical lenses of different imaging cameras of the continuous zoom optical system, and set time points of capturing images of different wave bands for the continuous zoom optical system to ensure that end time points of imaging of the continuous zoom optical system are the same.

[0057] A continuous zoom optical system includes a plurality of zoom lens control systems and a plurality of imaging cameras.

[0058] The imaging cameras include a visible light camera, a short wave camera, a medium wave camera, and a long wave camera.

[0059] The corresponding visible light camera, short wave camera, medium wave camera, and long wave camera are connected to the corresponding zoom lens control system to adjust the focal length of the corresponding camera.

[0060] Each zoom lens control system is connected to the imaging control system, receives parameter information of each imaging camera sent by the imaging control system to adjust the focal length of the imaging camera, and receives the set time points of capturing images of different wave bands to ensure that the end time points of imaging of the continuous zoom optical system are the same.

[0061] The image analysis processing system includes an image preprocessing module, an image stitching module, a target detection and recognition module, an analysis control module, and a video output module.

[0062] The image preprocessing module is configured to preprocess images of different wave bands received from the continuous zoom optical system, and send the preprocessed images of different wave bands to the image stitching module and the target detection and recognition module.

[0063] The image stitching module is configured to stitch the images of different wave bands into one picture and form a video stream, and output the video stream to the terminal in a set video format.

[0064] The target detection and recognition module is configured to perform target recognition on the preprocessed images of different wave bands, calculate result information of the target, and send the preprocessed images and the result information to the analysis control module.

[0065] The analysis control module is configured to analyze the result information to obtain parameter information of each imaging camera in the continuous zoom optical system, and send the parameter information back to the imaging control system for real-time control.

[0066] The pre-processed images of the same waveband are tracked by a gate tracking method or a feature matching method, and the image stream in the tracking process is sent to a video output module;

[0067] The pre-processed images of different wavebands are fused by an HSV fusion algorithm or a contour fusion algorithm, and the image stream in the fusion process is sent to a video output module;

[0068] The video output module is configured to output in a set format according to the selection of the analysis control module.

[0069] The result information includes target size, target motion direction and target motion rate, and the parameter information includes zoom lens focal length size, focal length step and frequency, and imaging camera shutter, gain setting value and adjustment direction.

[0070] The imaging control system comprises an imaging control module and an image synchronization control module connected with each imaging camera of the continuous zoom optical system respectively;

[0071] The imaging control module is configured to receive the parameter information of each imaging camera in the continuous zoom optical system sent by the analysis control module, and control according to the parameter information of each imaging camera;

[0072] The image synchronization control module is configured to set the image video acquisition time points of each imaging camera in the continuous zoom optical system, so as to ensure that the ending time points of imaging are the same.

[0073] As shown in Figure 1 , it is a structural schematic diagram of the present application, the continuous zoom optical system and the image analysis processing system and the imaging control system connected therewith are integrated together; the continuous zoom optical system is used in each imaging system, the zoom mode is cam zoom, the optical axis consistency is good, the reliability is high, and the control is simple. The other functional modules of the continuous zoom optical system in the system are arranged on an image processing board (including FPGA and DSP) to be realized, and other CPUs can also be selected to be realized;

[0074] As shown in Figure 2 , it is a system block diagram and method flow schematic diagram of the present application, the present application provides a processing method of a portable multi-spectral video image recording and analysis processing system, comprising the following steps:

[0075] 1) The continuous zoom optical system collects images of different wavebands of a target, and sends the images to an image preprocessing module;

[0076] 2) The image preprocessing module pre-processes the images of different wavebands, and sends the pre-processed images to an image splicing module and a target detection and recognition module respectively;

[0077] 3) image stitching module by multi-band stitching method to the same time different waveband image stitching into a picture and form a video stream, and set the unified resolution output to the video stream storage terminal;

[0078] At the same time, the target detection and recognition module for different waveband preprocessed image target recognition, and calculate the target result information, the result information and the target feature extracted to the preprocessed send to the analysis control module;

[0079] 4) analysis control module according to the result information analysis, draw the continuous zoom optical system in each imaging camera parameter information, back to the imaging control system for real-time control of the continuous zoom optical system in each imaging camera;

[0080] At the same time, the analysis control module for the same waveband preprocessed image using gate tracking method or feature matching method are tracked, the image stream in the tracking process is sent to the video output module;

[0081] Analysis control module for different waveband preprocessed image using HSV fusion algorithm or contour fusion algorithm are fusion processing, the image stream in the fusion process is sent to the video output module;

[0082] 5) video output module for analysis control module selection, output in the set format.

[0083] Image preprocessing module for different waveband image preprocessing, specifically:

[0084] The images collected by each imaging camera are sequentially denoised, filtered, image enhanced, edge extracted, and fog removed to obtain multiple different waveband preprocessed images.

[0085] Target detection and recognition module, different waveband preprocessed image target recognition, specifically:

[0086] (1) image preprocessing module to obtain different waveband secondary image respectively applied foreground background segmentation method, target detection and recognition module to obtain color, gray value, gradient, histogram target feature, while the background information is segmented out;

[0087] (2) after detection and recognition to the target, calculate the target size, target motion direction, target motion rate.

[0088] According to the target feature, the target size, target motion direction and target motion rate are obtained as follows:

[0089] The target feature is extracted by using the background segmentation method. First, a background model is established, and the feature parameters of the image sequence are compared to segment the background and foreground, thereby obtaining the target area and edge, and further obtaining the target size.

[0090] Meanwhile, the color, gray value, gradient and histogram target feature of the target are obtained. The target motion direction can be determined by the position of the target in the image in multiple frames of images, and the target motion rate is obtained by combining the time of each frame of image.

[0091] In step 4), the analysis control module analyzes the result information to obtain the parameter information in each imaging camera in the continuous zoom optical system, specifically:

[0092] According to the target size, the focal length adjustment size of the zoom lens is obtained to ensure that the target size is maintained within the set range at the focal length, that is, the target size is increased when the focal length is increased, and the target size is reduced when the focal length is reduced.

[0093] According to the target motion direction, target motion rate and target size, the step size and frequency of changing the focal length are determined, that is, the larger the target or the larger the target motion rate, the larger the step size and the higher the frequency of changing the focal length, and vice versa, the smaller the target or the smaller the motion rate, the smaller the step size and the lower the frequency of changing the focal length.

[0094] According to the brightness of the target, the average gray value of the background and the histogram, the shutter, gain setting value and adjustment direction of the corresponding imaging camera are determined.

[0095] Embodiment:

[0096] In the portable multi-spectral video recording stage: the continuous zoom optical system is used in the visible light, short wave, medium wave, long wave and other imaging cameras to adapt to the different optical lens focal lengths required by the target image video acquisition at different distances, and the parameter change control of the imaging system control module including the cameras in the visible light, short wave, medium wave, long wave and other imaging systems and the change of the focal length of each are used to collect images of different wave bands. At the same time, the image synchronous control module is used to control the image video acquisition time points of the visible light, short wave, medium wave, long wave and other imaging cameras to ensure that the end time points of their imaging are the same and can be traced back.

[0097] In this embodiment, each imaging camera is composed of a long-wave uncooled infrared lens, a long-wave uncooled core, a medium-wave cooled infrared lens, a medium-wave cooled infrared core, a television imaging system, a short-wave imaging system and a component frame, and the optical axis consistency is ensured to be less than or equal to 5 pixels. The continuous zoom optical system is used in each imaging system, and the zoom mode is cam zoom, which can ensure good optical axis consistency and high reliability, and the control is simple.

[0098] In the multi-spectral video image analysis processing stage:

[0099] The image preprocessing module refers to performing preprocessing such as denoising, filtering, image enhancement, edge extraction, and dehazing on visible light, short-wave, medium-wave, and long-wave imaging images according to the environmental conditions and prior information of the target during data acquisition, so as to obtain secondary images;

[0100] like Figure 3 The diagram shows a schematic of the image stitching module. In this embodiment, the resolutions of shortwave, mediumwave, and longwave in the multispectral images of different bands are all 640*512, with frame rates of 100Hz, 100Hz, and 50Hz, respectively; the resolution of visible light is 1920*1080, with a frame rate of 50Hz. The image stitching module stitches the visible light, shortwave, mediumwave, and longwave images into a single image in chronological order to form a video stream. Without loss of quality, it stitches the images into a 1080p50Hz image (containing two frames each of shortwave and mediumwave images, and one frame each of visible light and longwave), and then outputs it through the SDI interface. The visible light image in the stitched image has two cases: one is 1920*1080 reduced to 640*360; the other is a selected area of ​​640*512 in the complete image, the position of which is determined by the terminal host computer. The black area in the image can be used to display characters, which is convenient for video recording and analysis.

[0101] The target detection and recognition module processes each video image independently. After detecting and recognizing the target, it calculates the result information (including target size, direction of motion, speed of motion, etc.).

[0102] The analysis and control module uses the results from the target detection and recognition module to derive camera parameter information and focal length information for each imaging system in real time according to a specific preset comprehensive analysis method. This ensures that the image size of the target in each optical imaging system is within a suitable range. It obtains multispectral image information of the target in different bands under various conditions (e.g., changing multiple camera parameters: shutter speed, gain; zoom lens focal length; changing the algorithm selection and preprocessing level in preprocessing). Based on the target size, the zoom lens adjusts its focal length to ensure that the target size remains within a set range at a given focal length. That is, the focal length increases when the target is small and decreases when the target is large.

[0103] The step size and frequency of changing the focal length are determined based on the target's direction of motion, speed of motion, and size. That is, the larger the target or the faster the target moves, the larger the step size and the higher the frequency of changing the focal length. Conversely, the smaller the target or the slower the target moves, the smaller the step size and the lower the frequency of changing the focal length.

[0104] Based on the target's brightness, the average grayscale value of the background, and the histogram, determine the shutter speed, gain setting, and adjustment direction of the corresponding imaging camera.

[0105] Meanwhile, the process images of the fusion algorithm and the tracking algorithm are sent to a video synthesis output module; the video output module is selected according to the analysis control module to output in a set format.

[0106] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A portable multispectral video image recording and analysis processing system, characterized in that The application relates to a continuous zoom optical system and an image analysis processing system and an imaging control system connected with the continuous zoom optical system. The continuous zoom optical system comprises multiple imaging cameras, is used for receiving parameter information controlled by the imaging control system, and provides optical lens focal lengths of different imaging cameras required by a target at different distances, simultaneously collects images of different wave bands, and sends the images to the image analysis processing system for processing. The image analysis processing system is used for receiving images of different wave bands collected by the continuous zoom optical system for analysis and processing, acquiring parameter information of the continuous zoom optical system, and sending the parameter information to the imaging control system, and simultaneously outputting to a terminal in a set format. The image analysis processing system comprises an image preprocessing module, an image splicing module, a target detection and recognition module, an analysis control module and a video output module. The image preprocessing module is used for preprocessing images of different wave bands received by the continuous zoom optical system, and sending the preprocessed images of different wave bands to the image splicing module and the target detection and recognition module respectively. The image splicing module is used for splicing images of different wave bands into one picture and forming a video stream, and outputting to a terminal in a set video format. The target detection and recognition module is used for performing target recognition on the preprocessed images of different wave bands, calculating result information of the target, and sending the preprocessed images and the result information to the analysis control module. The analysis control module is used for performing analysis according to the result information, obtaining parameter information of each imaging camera in the continuous zoom optical system, and feeding back to the imaging control system for real-time control. The preprocessed images of the same wave band are tracked by using a gate tracking method or a feature matching method respectively, and image streams in a tracking process are sent to the video output module. The preprocessed images of different wave bands are fused by using an HSV fusion algorithm or a contour fusion algorithm respectively, and image streams in a fusion process are sent to the video output module. The video output module is used for outputting in a set format according to selection of the analysis control module. The imaging control system is used for receiving parameter information of the continuous zoom optical system of the image analysis processing system, and controlling optical lens focal lengths of different imaging cameras of the continuous zoom optical system, and simultaneously setting image collection time points of different wave bands of the continuous zoom optical system, so as to ensure that ending time points of imaging of the continuous zoom optical system are the same. The continuous zoom optical system comprises multiple zoom lens control systems and multiple imaging cameras.

2. A portable multispectral video image recording and analysis processing system according to claim 1, characterized in that, The imaging cameras comprise visible light cameras, short wave cameras, middle wave cameras and long wave cameras. The visible light cameras, the short wave cameras, the middle wave cameras and the long wave cameras are connected with corresponding zoom lens control systems, so as to adjust lens focal lengths of the corresponding cameras. ​ Each zoom lens control system is connected with the imaging control system, and by receiving parameter information of each imaging camera sent by the imaging control system, the lens focal length of the imaging camera is adjusted, and the collection time points of images of different wave bands are received, so that the ending time points of continuous zoom optical system imaging are ensured to be the same.

3. A portable multispectral video image recording and analysis processing system according to claim 1, characterized in that, The result information includes target size, target motion direction, and target motion rate. The parameter information includes zoom lens adjustment focal length size, focal length change step and frequency, and imaging camera shutter value, gain setting value, and adjustment direction.

4. A portable multispectral video image recording and analysis processing system according to claim 1, characterized in that, The imaging control system includes an imaging control module and an image synchronization control module connected with each imaging camera of the continuous zoom optical system respectively. The imaging control module is configured to receive parameter information of each imaging camera in the continuous zoom optical system sent by the analysis control module, and control according to the parameter information of each imaging camera. The image synchronization control module is configured to set image video collection time points of each imaging camera in the continuous zoom optical system, so that the ending time points of imaging are ensured to be the same.

5. The processing method of a portable multispectral video image recording and analysis processing system according to claim 1, characterized in that, The method includes the following steps: 1) The continuous zoom optical system collects images of different wave bands of a target, and sends the images to an image preprocessing module; 2) After the image preprocessing module pre-processes the images of different wave bands, the images are sent to an image splicing module and a target detection and recognition module respectively; 3) The image splicing module splices the images of different wave bands at the same time into one picture by a multi-wave band splicing method, forms a video stream, and outputs the video stream to a video stream storage terminal at a set unified resolution; Meanwhile, the target detection and recognition module performs target recognition on the pre-processed images of different wave bands, calculates result information of the target, and sends the result information and extracted target features after pre-processing to an analysis control module; 4) The analysis control module analyzes the result information, obtains parameter information of each imaging camera in the continuous zoom optical system, and returns the parameter information to the imaging control system to control each imaging camera in the continuous zoom optical system in real time; Meanwhile, the analysis control module performs tracking processing on the pre-processed images of the same wave band by a gate tracking method or a feature matching method, and sends image streams obtained in the tracking process to a video output module; The analysis control module performs fusion processing on the pre-processed images of different wave bands by an HSV fusion algorithm or a contour fusion algorithm, and sends image streams obtained in the fusion process to the video output module; 5) The video output module outputs in a set format according to the selection of the analysis control module.

6. The processing method of a portable multispectral video image recording and analysis processing system according to claim 5, characterized in that, The image preprocessing module pre-processes the images of different wave bands, specifically: The images collected by each imaging camera are sequentially subjected to denoising, filtering, image enhancement, edge extraction, and fog removal processing, to obtain multiple pre-processed images of different wave bands.

7. The processing method of a portable multispectral video image recording and analysis processing system according to claim 5, characterized in that, In step 3), the target detection and recognition module performs target recognition on the pre-processed images of different wave bands, specifically: (1) The image preprocessing module obtains secondary images of different wave bands, and the target detection and recognition module obtains target features of color, grayscale value, gradient, and histogram, while the background information is segmented out by applying a foreground and background segmentation method. (2) After detecting and identifying the target, the target size, the target motion direction and the target motion rate are calculated.

8. The processing method of a portable multispectral video image recording and analysis processing system according to claim 7, characterized in that, The method for obtaining the target size, the target motion direction and the target motion rate according to the target features is as follows: The target features are extracted by using the background segmentation method. First, a background model is established and compared with the feature parameters of the image sequence to segment the background and the foreground, thereby obtaining the target area and the edge, and further obtaining the target size. Meanwhile, the color, the gray value, the gradient and the histogram target features of the target are obtained. The target motion direction can be determined by the position of the target in the image in multiple frames, and the target motion rate can be obtained by combining the time of each frame.

9. The processing method of a portable multispectral video image recording and analysis processing system according to claim 5, characterized in that, In step 4), the analysis control module analyzes the result information to obtain the parameter information in each imaging camera in the continuous zoom optical system, specifically: The target size is used to obtain the focal length adjustment size of the zoom lens to ensure that the target size is kept within the set range at the focal length, i.e., the focal length is increased when the target size is decreased, and the focal length is decreased when the target size is increased; The target motion direction, the target motion rate and the target size are used to determine the step size and the frequency of changing the focal length, i.e., the larger the target or the target motion rate, the larger the step size and the higher the frequency of changing the focal length, and vice versa; The brightness of the target, the average gray value of the background and the histogram are used to determine the shutter, the gain setting value and the adjustment direction of the corresponding imaging camera.

Images