A real-time image tracking performance test system

Abstract

The application discloses a real-time image tracking performance test system, which comprises a host computer, an instruction interface conversion circuit, a switching gate control module, an image source generator, a video format interface circuit and a tracking performance processing module; the performance test instruction sent by the host computer is directly sent to the tracking performance processing module after being processed by the switching gate control module; the tracking performance processing module controls the image source generator to send image frames and coordinate values matched with the instruction, and controls the internal timer to work, so that the measured image tracker performs corresponding operation and returns test data to the tracking performance processing module; after receiving the measured data and the test data, the tracking performance processing module calculates tracking performance test results and feeds back the results to the host computer.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology and relates to a real-time image tracking performance testing system. In particular, it relates to a testing system for the tracking performance of an image tracker (test device) product in a guided weapon system that can accurately strike targets, relocate and track targets that disappear in a short time, and track targets under low contrast conditions. The system includes performance tests for dynamic tracking accuracy, static tracking accuracy, manual tracking accuracy, tracking response time, and search and reacquisition time. Background Technology

[0002] Image trackers can automatically identify static or dynamic target images, enabling manual or automatic target positioning and tracking. They are widely used in military and civilian image processing fields, such as target acquisition and engagement in weapon systems, target capture and photography by drones, and camera recording. In image tracking systems for precision strikes or target acquisition, environmental factors such as wind, birds, and sunlight, especially noise, jitter, product malfunctions, or defects in the image tracker itself, can affect tracking accuracy, leading to inaccurate target tracking. Furthermore, the accuracy of target tracking cannot be determined by the human eye and measured within tens of milliseconds or seconds. Therefore, highly reliable methods for real-time image tracking performance testing are crucial and can be applied to various image trackers to ensure the stable and reliable operation of image tracking systems.

[0003] Available domestic and international literature describes how to test the tracking accuracy of such image tracking systems, such as Chinese patent No. 201510566567.X, "A Tracking Accuracy Inversion Method Based on Sequence Images." This patent introduces the steps of a tracking accuracy inversion method based on sequence images: first, based on the known orbit / attitude data of the observed satellite, the orbit / attitude data of the target satellite, and a continuous image sequence, a coordinate transformation method is used to obtain the ideal position of the selected target point; second, the actual position of the target point is obtained by searching for the basic transformation matrix based on the same-plane feature points; third, the difference between the ideal position and the actual position of the target point is calculated, and combined with the given camera parameters, the inversion error expressed in angle form can be obtained. However, this patent does not have the ability to transform the target point in real time, nor can it perform real-time image tracking accuracy testing and provide test results.

[0004] Chinese patent No. 202010919713.3, entitled "A Multi-Channel Video Synchronous Switching System and Method Based on FPGA," describes a system comprising an arbitration module, a video matrix module, two frame synchronization modules, and a video synchronization switching module. When N video streams need to be synchronously switched, this method dynamically configures the DDR access permission for each video stream based on the currently displayed video source and the target video source to be switched. Only the currently displayed and soon-to-be-displayed video streams are allowed to access the DDR, while idle video streams are prevented from occupying the DDR, thus improving memory utilization during multi-channel video synchronization. However, this patent lacks the ability to generate multiple video or image sources, and it also lacks the capability to test image tracking accuracy, tracking response time, etc.

[0005] Patent No. 201711343812.6, entitled "High-Precision Infrared Image Tracker and a Fast Target Tracking Method," discloses a high-precision infrared image tracker and a fast target tracking method. This tracker is based on a programmable logic array and a digital processing chip framework, including a real-time image acquisition and processing module, a real-time image storage module, a control module (FPGA), an output image storage module, a communication module, a digital processing module (DSP), and a dynamic storage module (SDRAM). This patent focuses on how to quickly track a target, but does not include details on how to detect the accuracy of target tracking or the tracking response time.

[0006] Currently, a patent novelty search has not revealed a system similar to this patent that can provide real-time image motion targets for an image tracker and test the tracking performance of the image tracker. Summary of the Invention

[0007] (I) Purpose of the Invention

[0008] The purpose of this invention is to provide a real-time image tracking performance testing system that tests the tracking performance of an image tracker product, including static / dynamic / manual tracking accuracy, tracking response time, and search-recapture time, without altering the product itself.

[0009] (II) Technical Solution

[0010] To solve the above-mentioned technical problems, the present invention provides a real-time image tracking performance testing system, characterized in that it includes: a host computer, an instruction interface conversion circuit, a switching gating control module, an image source generator, a video format interface circuit, and a tracking performance processing module;

[0011] The host computer issues a performance test command for the image tracker according to the operator's instructions, waits for the underlying system to automatically complete the performance test, and then receives and displays the performance test results.

[0012] The instruction interface conversion circuit converts a performance test instruction sent by the host computer into a performance test signal that can be received by the switching gating control module;

[0013] The switching gating control module sends a serial control signal to the image source generator and a performance test serial control signal to the tracking performance processing module based on the content of a certain performance test signal.

[0014] The image source generator produces image sources of various controllable moving targets, and outputs the coordinate values ​​of the center point of the moving target contained in the image frame;

[0015] When the image source generator receives a serial control signal, it sends out an image source of a specific moving target for the image tracker to perform automatic or manual image tracking.

[0016] Upon receiving an image signal, the image tracker immediately performs automatic or manual tracking and actively outputs the coordinates of the center point of the tracked moving target. The tracking performance processing module acquires the original center point coordinates of the moving target in each frame of the image source generator and the center point coordinates of the tracked moving target output by the image tracker in real time. It also calculates the tracking accuracy of each frame by rapidly calculating the difference between the two coordinates in real time. The tracking performance processing module ensures that the time signatures of the two coordinate information are consistent, performs effective comparison of the two coordinate values, and finally outputs the tracking accuracy performance test results for different detection time ranges.

[0017] This invention also provides a real-time image tracking performance testing method. The testing process is as follows: the performance test command sent by the host computer is processed by the switching gating control module and then directly sent to the tracking performance processing module. The tracking performance processing module controls the image source generator to emit image frames and coordinate values ​​that match the command, and controls the internal timer of the module to work, so that the image tracker under test performs corresponding operations and returns test data to the tracking performance processing module. After receiving the test data and the test data, the tracking performance processing module calculates the tracking performance test result and feeds it back to the host computer.

[0018] (III) Beneficial Effects

[0019] The real-time image tracking performance testing system provided by the above technical solution has the following beneficial effects.

[0020] (1) This invention abandons the previous practice that image source generators could only emit a single image source and the moving targets were not diverse enough. Instead, it adopts a variety of image sources and can select a certain image source to output a specific moving target image according to the test requirements of the image tracker.

[0021] (2) The present invention abandons the previous practice of using real-time camera shooting as the image source and avoids the following disadvantages: 1) Real-time camera shooting cannot provide accurate target coordinate information; 2) It is difficult to provide low-contrast images; 3) It is also difficult to simulate the effect of the target disappearing temporarily.

[0022] (3) One existing method for testing tracking accuracy is to continuously collect the tracking point coordinate information output by the image tracker within a fixed detection time, and after the collection is completed, use software to draw the tracking point coordinate information into a curve and fit it with the original motion target curve of the image source generator, and give the tracking accuracy result of each coordinate point after fitting. This invention abandons the existing testing methods that have no selectable detection time, poor real-time performance, and large tracking accuracy test errors.

[0023] (4) This invention can effectively detect whether the static / dynamic / manual tracking accuracy of a guided weapon image tracker meets the system index requirements through rapid data format conversion and statistical criteria. It can realize real-time monitoring of image tracker accuracy and automatic recording of output results.

[0024] (5) The present invention can effectively detect the search and recapture time value of an image tracker by simulating the image source of a moving target that disappears briefly and then reappears.

[0025] (6) The present invention can effectively detect the tracking response time of the image tracker by accurately obtaining the start time of the moving target in the image frame being tested.

[0026] (7) It features no missed detection of tracking pixels, real-time detection of product tracking accuracy, and selectable tracking accuracy test time range.

[0027] (8) The present invention has the characteristics of high real-time performance, compatibility with multiple communication interfaces, high reliability and low cost. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the principle components of a real-time image tracking performance testing system.

[0029] Figure 2 This is the program flowchart for switching the gating control module.

[0030] Figure 3 This is a schematic diagram of the image source generator.

[0031] Figure 4 It is a standard trajectory diagram of the target for tracking accuracy testing.

[0032] Figure 5 This is a schematic diagram showing the coordinates of the center point of the moving target and the coordinates of its movement range.

[0033] Figure 6 This is the program flowchart for the moving target x calculation module.

[0034] Figure 7 This is the program flowchart for the CML format encoding module.

[0035] Figure 8 This is a flowchart of the program for dynamic / static / manual tracking accuracy testing of the tracking performance processing module.

[0036] Figure 9 This is a flowchart of the program used by the performance processing module to track response time and search / recapture time tests. Detailed Implementation

[0037] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0038] This invention addresses the shortcomings of current high-precision image trackers used in guiding weapon systems for accurately tracking targets by providing a real-time performance testing system with selectable testing time range. This system can perform five types of performance tests, including dynamic tracking accuracy and tracking response time.

[0039] The schematic diagram of the high-reliability real-time image tracking performance testing system is attached. Figure 1 As shown, it includes a host computer, an instruction interface conversion circuit, a switching gating control module (embedded MCU chip superimposed with CPLD chip), an image source generator (FPGA1), a video format interface circuit, and a tracking performance processing module (DSP1).

[0040] The host computer issues a performance test command for the image tracker based on the operator's instructions, waits for the underlying system to automatically complete the performance test, and then receives and displays the performance test results.

[0041] The instruction interface conversion circuit converts a performance test instruction sent by the host computer into a performance test signal that the switching gating control module can receive. The switching gating control module then, based on the content of the performance test signal, sends both a serial control signal to the image source generator and a performance test control serial signal to the tracking performance processing module.

[0042] The image source generator can generate image sources for various controllable moving targets, and can also output the coordinates of the center point of the moving target contained in the image frame. The video format of the image source varies depending on the video format required by the image tracker under test, including but not limited to CML format, and can also be Cameralink format, HDMI format, SDI format, DVI format, VGA format, etc. After receiving a certain serial control signal, the image source generator sends out the image source of the specific moving target for the image tracker to perform automatic or manual image tracking, and the correspondence should follow the table below.

[0043] Table 1. Correspondence between performance test commands and image sources

[0044]

[0045] Upon receiving an image signal, the image tracker immediately performs automatic or manual tracking and actively outputs the coordinates of the center point of the tracked moving target. The tracking performance processing module acquires the original center point coordinates of the moving target from the image source generator and the tracked moving target center point coordinates from the image tracker in real time. It also calculates the tracking accuracy for each frame by rapidly calculating the difference between the two coordinates. The tracking performance processing module ensures that the time signatures of the two coordinate information values ​​are consistent, enabling effective comparison and ultimately outputting tracking accuracy performance test results for different detection time ranges such as 10s and 20s.

[0046] The aforementioned testing system includes the following important modules, which are described in detail below:

[0047] (1) Design a switching gating control module. The program flowchart of the module is as follows. Figure 2 As shown. Its hardware can be implemented using an embedded MCU chip superimposed with a CPLD chip, requiring multiple serial data interfaces. The switching control module is mainly used to receive performance test commands sent by the host computer, parse the commands, and immediately execute the gating function to select the serial ports required for performance testing, enabling certain serial ports to communicate with each other, such as selecting the serial port for transmitting center point coordinate values ​​between the image source generator and the tracking performance processing module. In addition, this module not only needs to send serial control signals to command the image source generator to emit appropriate moving target images, but also needs to send performance test signals to the tracking performance processing module.

[0048] (2) Design an image source generator, the hardware of which is composed of an FPGA. The image source generator can be controlled by different control signals (this patent only uses the UART format image source generator control signal as an example) to generate CML format image frames (image size 1000×1000 pixels), which contain specified moving targets. For example... Figure 3 The diagram shown is a schematic of the components of an image source generator.

[0049] The serial signal processing module processes the control signal and outputs the pattern variable.

[0050] The motion target selection module receives the pattern variable and generates the corresponding enable variable enable, enabling the corresponding motion target calculation module. On the other hand, it receives the coordinate values ​​of the center point of the motion target output by the enabled motion target calculation module N (N=1~9), converts them into UART format, and outputs them to the tracking performance processing module.

[0051] The moving target calculation module includes, but is not limited to, the following image sources 1-9, wherein the motion trajectories of signal sources 1-6 are as follows: Figure 4 As shown, the corresponding description is as follows:

[0052] a) Image source 1: The image contains a moving target with a size of (7×7) pixels, a contrast of no more than 3%, and a motion speed of 2 pixels / frame. The target disappears for 1 second during the motion.

[0053] b) Image source 2: The image contains a target (azimuth, elevation) with a size of (20×20) pixels, a contrast of no more than 3%, and a motion speed of 2 fields of view / second;

[0054] c) Image source 3: The image contains a moving target with a size of (20×20) pixels, a roll angle range of +30° to -30°, and an absolute roll angular velocity of 80° / s;

[0055] d) Image source 4: The image contains a target with feature points that grows from small (initial size of (7×7) pixels with a contrast of no more than 3%) to fill the field of view (this change in the size of the target simulates the change in the size of the target in the field of view as the missile approaches the target from 2 km).

[0056] e) Image source 5: The image contains a stationary target with a size of (7×7) pixels and a contrast of no more than 3%;

[0057] f) Image source 6: The image contains a moving target with a size of (7×7) pixels, a contrast of no more than 3%, and a motion speed of 2 pixels / frame.

[0058] g) Image source 7: The image contains a moving target with a size of (4×4) pixels and a motion speed of 2 pixels / frame;

[0059] h) Image source 8: The television image contains a target that grows from small to large, starting at (7×7) pixels and increasing to (800×800) pixels;

[0060] i) Image source 9: The infrared image contains a target that grows from small to large, starting at (7×7) pixels and increasing to (500×400) pixels.

[0061] This embodiment adopts, but is not limited to, moving targets being squares of varying sizes, such as... Figure 5 As shown. The coordinates of the center point of the square are the coordinates of the center point of the moving target. The side length of the square is determined by the size of the moving target, and the coordinate range of the moving target can be calculated from it.

[0062] The number of moving target calculation modules is determined by the number of image sources. This patent designs a total of 9 moving target calculation modules. Taking any one of them as an example, the flowchart of the moving target x calculation module is as follows: Figure 6 As shown, it can calculate and output the center point coordinates Xi, Yi and the coordinate range values ​​Xi1 and Yi1, Xi2 and Yi2, Xi3 and Yi3, Xi4 and Yi4 of a moving target in real time, where i = (0...N). This patent uses... Figure 3 Taking the trajectory as an example, the center point coordinates Xi and Yi, and the coordinate range value Xi 1～4 Yi 1～4 The following formulas (1) to (10) can be used for calculation:

[0063] Xi+1=Xicosθ(θ=0°~180°) (1)

[0064] Yi = Xi (2)

[0065] Xi1=Xi-a / 2 (3)

[0066] Yi1=Yi+a / 2 (4)

[0067] Xi² = Xi + a / 2 (5)

[0068] Yi2=Yi+a / 2 (6)

[0069] Xi3=Xi+a / 2 (7)

[0070] Yi3=Yi-a / 2 (8)

[0071] Xi4=Xi-a / 2 (9)

[0072] Yi4=Yi-a / 2 (10)

[0073] The flowchart of the moving target x calculation module describes how the program first determines the trajectory formula of the image target based on the known conditions of the image source; then, it determines the side length of the square representing the image target, thereby calculating the coordinate range of the moving target, i.e., the coordinate values ​​of the four vertices of the square; finally, it determines the moving target's X and Y velocities Vx and Vy in the X and Y directions. After the program enters an idle state, the output valid variable doenble is cleared to zero, and it waits for the enable variable enable to be selected as x, and the encoding valid variable moveeon to be 1. Once both conditions are met, the coordinate values ​​and coordinate range of the moving target's center point from the starting frame to the next frame are calculated. After the calculation is complete, the output valid variable doenble is set to 1, ensuring that the coordinate values ​​and coordinate range of the moving target's center point in each frame can be effectively received and processed by the CML format encoding module.

[0074] The flowchart of the CML format encoding module is as follows: Figure 7 As shown, it can encode the coordinates of the center point of a moving target in real time and output parallel video signals in CML format. After initialization, once the output valid variable doenable is set to 1 by the selected moving target calculation module, it first sends a high-level control signal TX_en and a frame header synchronization signal to the video encoding interface chip, and then outputs the encoding valid variable moveon = 1 to the moving target calculation module, enabling the moving target calculation module to calculate and output the center point coordinates and coordinate range of each frame. After sending a 16-bit wide image information packet to the video encoding interface chip, the module waits for and receives the pixel coordinate values ​​returned by the moving target calculation module. It judges the pixel coordinate values, encodes the image according to different gray levels based on the coordinate value range, and outputs line valid data, line blanking overlay data packets, and frame blanking data packets. After the encoding information of lines 1 to 1000 of each frame is sent, it sends a low-level control signal TX_en = 0 to the video encoding interface chip and sets the output valid variable doenable to 0.

[0075] (3) Design a tracking performance processing module to receive different performance test commands in real time. The module software operates in different modes according to the performance test commands, and in the appropriate operating mode, it interrupts and interacts with the image source generator, continuously receiving the center point coordinates of the moving target and the center point tracking coordinates output by the image tracker, and finally returns the performance test results to the host computer within a specified time.

[0076] Based on a real-time image tracking performance testing system, this invention includes a method for testing tracking accuracy. The testing process is as follows: the performance test command sent by the host computer is processed by a switching gating control module and then directly sent to the tracking performance processing module. The tracking performance processing module controls the image source generator to emit image frames and coordinate values ​​matching the command, and controls the internal timer of the module to work, so that the image tracker under test performs corresponding operations and returns test data to the tracking performance processing module. After receiving the test data and the data under test, the tracking performance processing module calculates the tracking performance test result and feeds it back to the host computer.

[0077] The method described in this embodiment is applicable to testing dynamic, static, and manual tracking accuracy. The flowchart for the dynamic / static / manual tracking accuracy performance test procedure is shown below. Figure 8 As shown, the specific steps are as follows:

[0078] Step 1: The tracking performance processing module receives performance test instructions (dynamic, static, or manual tracking accuracy performance test instructions);

[0079] Step 2: The tracking performance processing module informs the image source generator to send out the image frame being measured and its coordinate values;

[0080] Step 3: The tracking performance processing module receives and records the original coordinates of the center point of the moving target in each frame of the image given by the image source generator within a certain period of time via the serial port (Xi, Yi) (i = 1...n);

[0081] Step 4: The image tracker performs dynamic / static / manual tracking of the specified moving target emitted by the image source generator. Simultaneously, it sends tracking signals to the tracking performance processing module via a communication interface such as CAN. Assume the tracking coordinates of the moving target's center point in each frame of the image given within 30 seconds are (X″). i ,Y″ i (i = 1...n);

[0082] Step 5: The tracking performance processing module receives and records the tracking coordinates of the center point of the moving target in each frame of the image given by the image tracker within a certain time period;

[0083] Step Six: The tracking performance processing module determines whether data format protocol conversion is involved. If so, it performs a rapid data format protocol conversion to facilitate the tracking error calculation in Step Seven. The manual tracking accuracy performance test of this invention involves the following data format protocol conversion process:

[0084] 1) The coordinates of the center point of the tracked moving target output by the image tracker are CAN data, and its data protocol is as follows:

[0085] a) Method: CAN bus communication baud rate 500Kbps (Timer 0: 0x00, Timer 1: 0x1C); Acceptance code: 00000000; Mask code: FFFFFFFF; Filtering method: Single filter; Mode: Normal mode; Data format: Extended frame; Transmission period: 20.8ms.

[0086] b) Data format:

[0087]

[0088]

[0089] Where k = 1, V max =7, F max =10 (as defined in this patent, but excluding the above protocol conversion formulas 13 and 14). V 输入 V represents the X or Y coordinates of the center point of the moving target in the original image. 输出 It is the X or Y coordinate value of the center point of the tracked moving target.

[0090] 2) The coordinates of the moving target's center point output by the image source generator's communication interface are UART data, and its communication data protocol is as follows:

[0091] a) Method: Serial, simplex communication; Communication format: 1 start bit, 8 data bits, 1 stop bit, even parity; Baud rate: 115200 bit / s; Transmission period: 2ms;

[0092] b) Data format: Consists of a header, body, and checksum; the data transmission format consists of 6 bytes (word1 to word6): word1: header, content is 0xaa; word2: the high 4 bits (right-aligned) of the X-axis coordinate value of the center point of the moving target in the original image; word3: the low 8 bits (right-aligned) of the X-axis coordinate value of the center point of the moving target in the original image; word4: the high 4 bits (right-aligned) of the Y-axis coordinate value of the center point of the moving target in the original image; word5: the low 8 bits (right-aligned) of the Y-axis coordinate value of the center point of the moving target in the original image; word6: checksum, which is the bitwise XOR value of word2 to word5.

[0093] 3) The tracking performance processing module needs to convert the coordinate values ​​of the center point of the moving target in the original image emitted by the image source generator into a data format in real time according to formulas (11 and 12) so that it is consistent with the data format of the coordinate values ​​of the center point of the moving target tracked by the image tracker.

[0094] Step 7: The tracking performance processing module calculates the average error of these n sets of measurements in real time, which is the tracking error. The specific formula is as follows:

[0095] △X i =|X i -X i "| and △Y i =|Y i -Y i "| (13)

[0096] Step 8: Control the tracking performance processing module to record the time for these two coordinate values, for example, set the time to 10s, 20s, 30s, etc.

[0097] Step 9: After the recording time ends, the tracking performance processing module will compare the tracking error of each frame in real time, perform statistical processing according to different tracking accuracy judgment criteria, and then output the final test results.

[0098] In this embodiment, the flowchart for the response time tracking test is as follows: Figure 9 As shown, the specific steps are as follows:

[0099] Step 1: The performance tracking module receives the performance test command for tracking response time;

[0100] Step 2: The tracking performance processing module sends an interrupt signal to inform the image source generator to emit image source 2 or image source 6, etc.

[0101] Step 3: The tracking performance processing module receives the time when the image source generator emits a frame of raw image containing a moving target, records it as the time start point t0, and starts the internal timer of the module to start timing;

[0102] Step 4: The tracking performance processing module receives the tracking coordinate values ​​of the frame image output by the image tracker at the time of the time endpoint t1; and shuts down the internal timer of the module to stop the timing.

[0103] Step 5: The performance tracking module calculates and outputs the tracking response time, using the following formula:

[0104] Δt=t1-t0(13);

[0105] In this embodiment, the flowchart for the search and recapture time test is as follows: Figure 9 As shown, the specific steps are as follows:

[0106] Step 1: The tracking performance processing module receives the search and recapture time performance test instruction;

[0107] Step 2: The tracking performance processing module sends an interrupt signal to inform the image source generator to emit an image source in which the moving target can automatically disappear during movement, such as image source 1 mentioned above;

[0108] Step 3: The tracking performance processing module receives a frame of original image from the image source generator. The moment when the moving target disappears for 1 second during the motion process and then reappears is recorded as the time start point t0; and the internal timer of the module is started to start timing.

[0109] Step 4: The tracking performance processing module receives the time when the tracking coordinates of the image frame reappear, output by the image tracker, and records it as the time endpoint t1; then it closes the internal timer of the module and stops timing.

[0110] Step 5: The tracking performance processing module calculates and outputs the search and recapture time, specifically using Formula 13.

[0111] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A real-time image tracking performance testing system, characterized in that, include: Host computer, instruction interface conversion circuit, switching and gating control module, image source generator, video format interface circuit, tracking performance processing module; The host computer issues a performance test command for the image tracker according to the operator's instructions, waits for the underlying system to automatically complete the performance test, and then receives and displays the performance test results. The instruction interface conversion circuit converts a performance test instruction sent by the host computer into a performance test signal that can be received by the switching gating control module; The switching gating control module sends a serial control signal to the image source generator and a performance test serial control signal to the tracking performance processing module based on the content of a certain performance test signal. The image source generator produces image sources of various controllable moving targets, and outputs the coordinate values ​​of the center point of the moving target contained in the image frame; When the image source generator receives a serial control signal, it sends out an image source of a specific moving target for the image tracker to perform automatic or manual image tracking. Upon receiving an image signal, the image tracker immediately performs automatic or manual tracking and actively outputs the coordinates of the center point of the tracked moving target. The tracking performance processing module acquires the original moving target center point coordinates of each frame of image output by the image source generator and the tracking moving target center point coordinates output by the image tracker product in real time. On the other hand, it calculates the tracking accuracy of each frame of image by quickly calculating the difference between the two coordinates in real time. The tracking performance processing module helps to ensure that the time stamps of the two coordinate information are consistent, performs effective comparison of the two coordinate values, and finally outputs the tracking accuracy performance test results for different detection time ranges. The image source generator includes a serial signal processing module, a moving target selection module, a moving target calculation module, and a CML format encoding module. The serial signal processing module processes the control signal and outputs a pattern variable. The moving target selection module receives the pattern variable and generates a corresponding enable variable. It also receives the moving target center point coordinates output by the moving target calculation module N, converts them to UART format, and outputs them to the tracking performance processing module. N = 1 to 9. The number of image sources determines the number of moving targets. The moving target calculation module calculates and outputs the moving target center point coordinates Xi, Yi, and coordinate range values ​​Xi1 and Yi1, Xi2 and Yi2, Xi3 and Yi3, Xi4 and Yi4 in real time, where i = 0...N. The CML format encoding module encodes the moving target center point coordinates within and outside the range in real time and outputs parallel video signals in CML format. The tracking performance processing module receives different performance test commands in real time, operates in different modes according to the performance test commands, and interacts with the image source generator in the appropriate mode. It continuously receives the center point coordinates of the moving target and the center point tracking coordinates output by the image tracker, and finally returns the performance test results to the host computer within a specified time.

2. The real-time image tracking performance testing system as described in claim 1, characterized in that, The video format of the image source varies depending on the video format required by the image tracker under test, including but not limited to CML format, Cameralink format, HDMI format, SDI format, DVI format, and VGA format.

3. The real-time image tracking performance testing system as described in claim 2, characterized in that, The switching gating control module is implemented by an embedded MCU chip superimposed with a CPLD chip and has multiple serial data interfaces. The switching gating control module receives performance test instructions sent by the host computer, parses the instructions and executes the gating function to select the serial port required for performance testing.

4. The real-time image tracking performance testing system as described in claim 3, characterized in that, The image source generator is composed of an FPGA. Different control signals control the image source generator to generate image frames in a set video format, which contain the required moving targets.

5. A method for testing real-time image tracking performance, characterized in that, The test method is based on the test system described in claim 4. The test process is as follows: the performance test command sent by the host computer is processed by the switching gating control module and then sent directly to the tracking performance processing module. The tracking performance processing module controls the image source generator to send out image frames and coordinate values ​​that match the command, and controls the internal timer of the module to work, so that the image tracker under test performs the corresponding operation and returns the test data to the tracking performance processing module. After receiving the data under test and the test data, the tracking performance processing module calculates the tracking performance test results and feeds them back to the host computer. The testing method for dynamic, static, or manual tracking accuracy testing follows these steps: Step 1: The tracking performance processing module receives dynamic, static, or manual tracking accuracy performance test commands; Step 2: The tracking performance processing module informs the image source generator to send out the image frame being measured and its coordinate values; Step 3: The tracking performance processing module receives and records the original coordinates of the center point of the moving target in each frame of the image given by the image source generator within a certain period of time (Xi, Yi), i=1……n; Step 4: The image tracker performs dynamic / static / manual tracking of the specified moving target emitted by the image source generator, and simultaneously sends a tracking signal to the tracking performance processing module through the communication interface. The tracking coordinates of the center point of the moving target in each frame of the image given within a set time period are ( , ), i=1……n; Step 5: The tracking performance processing module receives and records the tracking coordinates of the center point of the moving target in each frame of the image given by the image tracker within a set time period; Step Six: The tracking performance processing module determines whether data format protocol conversion is involved. If so, it performs a fast data format protocol conversion to facilitate the tracking error calculation in Step Seven. Step 7: The tracking performance processing module calculates the average error of these n sets of measurements in real time, which is the tracking error. The specific formula is as follows: and Step 8: Control the timing of the tracking performance processing module recording these two coordinate values; Step 9: After the recording time ends, the tracking performance processing module will compare the tracking error of each frame in real time, perform statistical processing according to different tracking accuracy judgment criteria, and then output the final test results. The testing method used for tracking response time testing follows these steps: Step 1: The performance tracking module receives the performance test command for tracking response time; Step 2: The tracking performance processing module sends an interrupt signal to inform the image source generator to emit an image source; Step 3: The tracking performance processing module receives the time when the image source generator emits a frame of raw image containing a moving target, records it as the time start point t0, and starts the internal timer of the module to start timing; Step 4: The tracking performance processing module receives the tracking coordinate values ​​of the original image of this frame output by the image tracker at the time, records it as the time end t1; and turns off the internal timer of the module to stop the timing. Step 5: The performance tracking module calculates and outputs the tracking response time, using the following formula: Δt = t1 - t0.

6. The real-time image tracking performance testing method as described in claim 5, characterized in that, The steps for testing the search and recapture time using the aforementioned testing method are as follows: Step 1: The tracking performance processing module receives the search and recapture time performance test instruction; Step 2: The tracking performance processing module sends an interrupt signal to the image source generator to emit an image source of the moving target that can automatically disappear during movement; Step 3: The tracking performance processing module receives a frame of original image from the image source generator. The moment when the moving target disappears for 1 second during the motion process and then reappears is recorded as the time start point t0; and the internal timer of the module is started to start timing. Step 4: The tracking performance processing module receives the time when the original image tracking coordinates of the frame reappear, output by the image tracker, and records it as the time endpoint t1; then it closes the internal timer of the module and stops timing. Step 5: The tracking performance processing module calculates and outputs the search and recapture time. The specific formula is as follows: Δt = t1 - t0.

Images