Target tracking and motion intention detection method and system based on color marking

Abstract

The invention belongs to the field of computer vision and human-computer interaction, and particularly relates to a target tracking and motion intention detection method and system based on color marking, which are used for target tracking and motion intention detection interactive design of an object. Comprising a target tracking method based on color marking and a motion intention enhancement detection method based on an average gray value. The method comprises the following steps: firstly, marking a target object by using a color mark, preprocessing an image through color segmentation and a mask, and visualizing by using an interpretable model; secondly, estimating the state of the system, and performing real-time target tracking in combination with Kalman filtering; in order to further improve the accuracy of target tracking, an image processing method of an average gray value is introduced, so that the influence caused by illumination variation is eliminated and the robustness of tracking is improved. And finally, evaluating the performance of the system and the model through image quality evaluation indexes and experimental research. Experimental results show that the method remarkably improves the structural definition, the signal-to-noise ratio and the equivalent number of looks of the image, shows high robustness in a complex scene, and achieves accurate tracking and stable motion intention detection in the complex scene.

Description

Technical Field

[0001] The present invention relates to a method and system for target tracking and motion intention detection based on color marking, in particular to a technical solution for real-time tracking of a target object and analyzing its motion intention in the field of computer vision and human-computer interaction. Background Art

[0002] With the rapid development of artificial intelligence and computer vision technology, target tracking and motion intention detection have become increasingly important in applications such as intelligent monitoring, autonomous driving, and virtual reality. Traditional target tracking methods usually rely on feature point matching and optical flow methods, but in complex scenes, factors such as lighting changes, occluded objects, and fast-moving targets can cause tracking accuracy to decrease, affecting the reliability and stability of the system.

[0003] At present, there are many research directions and technical methods in the field of target tracking and motion intention detection. These methods include intention perception planning methods for target tracking, which optimize tracking strategies by perceiving the target's motion intention; devices for identifying motion intentions, which use sensor data and algorithm analysis to identify the target's motion state and behavior; search and tracking methods based on intention reasoning and deep reinforcement learning, which combine deep learning technology to improve the intelligence level of target tracking; and interactive model state estimation methods based on intention inference, which improve the model's state estimation through user interaction feedback.

[0004] Although existing research has made some progress in target tracking and motion intention detection, the existing technology still has shortcomings in dealing with complex scenes, lighting changes, and real-time feedback. This makes it still challenging to achieve accurate and efficient target tracking and motion intention detection in dynamic environments. Therefore, the development of a new target tracking and motion intention detection method and system based on color marking has important practical significance and application value.

[0005] In recent years, color marker-based tracking technology has gradually attracted attention. Color markers have a sharp contrast and recognition in visual terms, making it easier to detect targets in various backgrounds. However, the existing color marker tracking system still has shortcomings in motion intention detection and cannot fully consider the target's motion trajectory, speed and other dynamic information, resulting in low accuracy in motion intention judgment.

[0006] In addition, the existing system is also relatively lacking in user interaction design, and it is difficult for users to obtain target status information in real time, which affects the human-computer interaction experience. Therefore, a new color-based target tracking and motion intention detection method and system is urgently needed, which can not only improve the accuracy of target tracking, but also enhance the detection ability of motion intention and provide a good user interaction experience.

[0007] The present invention aims to solve the above technical problems and proposes a target tracking and motion intention detection method and system based on color marking to improve the accuracy and robustness of target tracking, meet the application requirements in complex scenarios, and optimize the user interaction design to improve the overall performance of the system. Summary of the invention

[0008] One object of the present invention is to provide an intention-aware planning method for target tracking: this method analyzes the target's motion trajectory and environmental dynamics, combines machine learning and data mining techniques, to better perceive the target's potential intentions and improve the intelligence level of the tracking system.

[0009] Another object of the present invention is a device for identifying movement intention: This type of research is devoted to developing various hardware and software devices, using sensors and image recognition technology to realize automatic recognition of the target's movement intention. These devices can obtain the target's dynamic information in real time, so as to make timely responses.

[0010] Another object of the present invention is to combine the search tracking method of intention reasoning and deep reinforcement learning

[0011] By using deep learning models, especially reinforcement learning technology, the system can learn and optimize its tracking strategy through historical data in complex environments, infer the target's motion intention, and enhance the system's adaptability.

[0012] Another object of the present invention is an interactive model state estimation method based on intent inference: this method adjusts the system's understanding and response capabilities in real time through interaction with the user, combined with dynamic estimation of the target state, to achieve more efficient target tracking and intent detection.

[0013] The present invention not only improves the accuracy of target tracking, but also enhances the adaptability of the system in complex environments, providing new ideas and solutions for applications in the fields of intelligent monitoring, autonomous driving, and robotics. With the continuous advancement of technology, future target tracking systems will be more intelligent, able to better understand and predict the target's motion intentions, and achieve a wider range of applications.

[0014] The present invention provides a method and system for target tracking and motion intention detection based on color marking, which mainly includes:

[0015] A novel color-marking based target tracking method is proposed: combining the traditional color marking method with the Kalman filter algorithm to achieve smoother and more stable target tracking.

[0016] The motion intention enhancement detection method based on average gray value analyzes the average gray value of the image, eliminates the influence of illumination changes, and improves the detection accuracy of motion intention.

[0017] A motion intention detection method based on average gray value is proposed: the motion intention is identified and predicted by analyzing the gray value changes of pixels in video images, and an interpretable model (class activation mapping, CAM) is introduced to observe the interest points of the image.

[0018] The system integrates deep learning algorithms to improve target recognition accuracy and real-time response capabilities.

[0019] A simple and intuitive user interface is designed to facilitate user operation and settings.

[0020] The distribution of four color channels in the same image and the scatter plot after color space separation are discussed.

[0021] The effectiveness of the proposed method is verified through image quality evaluation and experimental research: the motion intention data is obtained by wearing IMU wearable devices through experimental design, and compared with the proposed method. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of a preferred embodiment of a system architecture of the present invention, including the target tracking module and the motion intention detection module.

[0023] Figure 2 It is the overall flow chart of the present invention, showing the process of system start-up and optimization.

[0024] Figure 3 The target tracking flowchart (left) and the motion intention detection algorithm flowchart (right) show the color marker extraction and tracking process and the motion intention analysis steps based on the average gray value.

[0025] Attached Figure 4 It is a user interaction interface diagram and a case analysis process diagram, showing the real-time data display and visualization of system feedback.

[0026] Figure 5 This is a specific implementation architecture diagram of the system, showing the overall detection and exception handling of target tracking and motion intention detection.

[0027] Figure 6 This is a simplified architecture diagram of the system for easy observation and analysis.

[0028] Component number description

[0029] 10 Color-based target tracking module

[0030] 20 Movement Intention Module 20

[0031] 30 Data acquisition module

[0032] 40 Data processing module

[0033] 50 System Operation Demonstration Module

[0034] 61~67 Target tracking module specific steps 71~77 Movement intention module specific steps 811~814 System status DETAILED DESCRIPTION

[0035] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only used as system examples for reference, and those skilled in the art can apply to other fields. The basic principles of the present invention defined in the following description can be applied to other implementation schemes, improved schemes, equivalent schemes, and other technical schemes that do not deviate from the spirit and scope of the present invention.

[0036] Those skilled in the art should note that, in the disclosure of the present invention, the terms "x-axis", "y-axis", "z-axis", "up", "down", "left", "right", "front", "back", "vertical", "horizontal", "move", "rotate", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, structure and operation. Therefore, the above terms should not be understood as limiting the present invention.

[0037] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0038] The present invention provides a target tracking and motion intention detection system, including a target tracking module 10 based on color marking, a motion intention module 20, a data acquisition module 30 and a data processing module 40 and a system practical demonstration module 50. The acquisition module 30 and the data processing module 40 can be connected in real time to collect and process data, wherein the data acquisition module 30 is connected to the target tracking module 10 and the motion intention module 20, so as to more accurately track and monitor the motion of the object, and the system is provided with corresponding early warning and alarm functions. Once the set threshold is exceeded, for example, a certain angle and speed are set, once exceeded, the system will issue a reminder and handle it abnormally.

[0039] The specific implementation steps and associations of the target tracking module 10 and the motion intention module 20 of the present invention are as follows: Figure 5The target tracking module 10 includes an initialization phase 61, a target tracking module 62, a motion intention detection module 63, data collection and processing 64, an early warning and alarm function 65, a practical demonstration module 66, and system feedback and optimization 67.

[0040] The initialization phase 61 refers to starting the system and loading the system configuration, including setting relevant parameters of the target tracking module, motion intention module, data acquisition module, data processing module and practical demonstration module; according to actual needs, setting the initial conditions for target detection, such as the recognition range of color markings, the direction and speed range of target movement.

[0041] The target tracking module 62 starts the color-based target tracking module 10, captures the scene image through a camera or sensor, and extracts the position information and motion trajectory of the target object according to the preset color mark. In this stage, the image data is processed in real time to extract the key features such as the position, speed, acceleration, etc. of the target; the tracked data is transmitted to the data acquisition module 30 in real time for further processing and analysis.

[0042] The motion intention detection module 63 analyzes the target's motion intention according to the target's motion trajectory and the preset model based on the motion intention module 20. This module needs to identify the target's motion trend (such as stop, accelerate, turn sharply, etc.); combine the target's motion speed, acceleration and change rate to determine whether there is abnormal motion, and transmit relevant data to the data acquisition module 30.

[0043] Data acquisition and processing 64 refers to the system data acquisition module 30 collecting data from the target tracking module and the motion intention module in real time and transmitting them to the data processing module 40; the data processing module 40 analyzes and fuses the collected multi-source data, including analysis of target motion, detection of abnormal motion and intention prediction, etc.; if motion is detected exceeding the set threshold (such as angle or speed exceeding a preset value), the system triggers an alarm mechanism.

[0044] The early warning and alarm function 65 is that when the system detects an abnormal situation (such as the target speed, acceleration, angle, etc. exceeding the set threshold), it automatically issues a warning and records relevant data; the system reminds the operator through sound, light, display screen, etc., and handles the abnormality. Emergency handling measures such as real-time tracking of abnormal targets, target stop or relocation can be set.

[0045] The practical demonstration module 66 is pre-processed by the practical demonstration module 50 and is used to demonstrate the operating effects and functions of the system. The system can be demonstrated through virtual demonstration or simulation training to help users better understand the working principles and operating procedures of the system; this module can combine the actual data of the target tracking module and the motion intention module for real-time demonstration and feedback.

[0046] System feedback and optimization 67 optimizes target tracking accuracy and motion intention prediction accuracy based on real-time data feedback; through user feedback and system self-learning, it gradually improves the early warning mechanism and enhances the system's adaptability to complex scenarios.

[0047] The specific implementation steps of the motion intention module 20 include: system setting 71, setting threshold 72, motion intention detection 73, system status information 74, system status signal 75, system abnormality detection 76, and output report 77.

[0048] The specific implementation steps of the motion intention module 20 are:

[0049] System Settings 71 refers to initializing the system configuration. When the system starts, it loads the configuration file of the motion intention module and sets the relevant parameters of the motion intention analysis, such as the maximum speed and acceleration of the target; sets the sensor parameters and adjusts the sensitivity of the sensor (such as accelerometer, gyroscope) according to different targets and environments to obtain more accurate motion data.

[0050] Set the threshold to 72. The system allows users to set the thresholds of the speed, acceleration and other parameters of the moving target according to actual needs. The setting of the threshold directly affects the subsequent intention detection and alarm functions; Threshold verification and adjustment:

[0051] During system operation, the system will automatically adjust the threshold according to the actual movement situation to ensure accurate prediction of the target behavior.

[0052] Motion intention detection 73 is divided into motion pattern detection. The system analyzes the target's motion data (such as position, speed, acceleration, etc.) to identify whether the target has normal or abnormal motion patterns; intention prediction and judgment is based on motion trajectory and historical data. The system predicts the target's next motion intention and determines whether the target may enter a dangerous state.

[0053] The system status information 74 is the motion status information of the target that is regularly updated by the system, including position, speed, acceleration, etc., to ensure the accuracy of the status information; the system transmits the updated status information to the data processing module in real time for further processing and analysis.

[0054] The system status signal 75 is a corresponding signal generated by the system according to the motion state of the target, such as normal state, warning state, alarm state, etc., for user reference; the system transmits the status signal to the control module and the alarm module for timely response.

[0055] System abnormality detection 76, the system continuously monitors the target's motion state and detects whether there is any abnormal behavior, such as speeding, emergency stop, etc.; when an abnormal state is detected, the system will trigger relevant emergency response mechanisms, such as target stop, alarm, etc.

[0056] Output report 77 is divided into report generation and report sending. The system generates detailed reports based on abnormal events, motion intention analysis results, etc. for subsequent analysis and decision-making; the reports are sent to relevant personnel via email, text messages, etc. to ensure real-time delivery of key information.

[0057] Through the above detailed implementation steps, the system can analyze and predict the movement intention while tracking the target in real time, and handle any abnormal situation in a timely manner. The close integration of the target tracking module and the movement intention module enables the system to work efficiently in complex environments, ensuring safety and reliability.

[0058] Specific implementation examples of the system: Example 1: In a monitoring scenario, the system tracks an athlete through color markings and predicts his next move (such as a sharp turn) based on the movement trajectory. If the athlete suddenly accelerates to the set threshold, the system immediately issues an alarm; Example 2: In an automated factory environment, the system tracks the movement of a transport trolley in real time. If the speed of the trolley exceeds the safety limit, the system triggers an alarm and automatically slows down.

[0059] The specific implementation of the present invention is simplified as follows. Figure 6 .

[0060] Step 1: Use a camera to capture the scene image, mark the target object with color, and select a color with good contrast as the mark.

[0061] Step 2: Apply image processing algorithms to perform color segmentation, extract target objects, and establish a color model of the target.

[0062] Step 3: Estimate the target's motion state through Kalman filtering to achieve real-time tracking. At the same time, the image is processed based on the average gray value method to eliminate the influence of illumination changes and enhance the detection ability of motion intention.

[0063] Step 4: The system analyzes the target's future motion intentions based on its motion trajectory and speed through a motion detection algorithm, and dynamically updates it based on historical motion data.

[0064] Step 5: The current position and motion status of the target are fed back in real time through the user interface, providing intuitive operation guidance.

[0065] Step 6: Users can adjust settings through the touch screen or gesture control. The system also supports voice commands to improve interaction convenience.

[0066] Step 7: Optimize the motion intention detection model based on the average grayscale value, and improve the accuracy and robustness of the model by enhancing the algorithm.

[0067] Through the above implementation steps, the color-marking-based target tracking and motion intention detection method and system of the present invention achieve accurate tracking and stable motion intention detection in complex scenarios, greatly improving the user experience and the overall performance of the system.

[0068] The protection scope of the target tracking and motion intention detection system and method described in the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0069] In summary, the present invention effectively overcomes the shortcomings of the prior art such as low target detection accuracy and high cost and has high industrial utilization value.

[0070] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only for illustrating the principles and effects of the present invention, and are not intended to limit the present invention. Any person familiar with the art may modify or alter the above embodiments without violating the spirit and scope of the present invention. Therefore, the embodiments of the present invention may be modified or altered in any manner without departing from the principles described above.

Claims

1. A method and system for target tracking and motion intention detection based on color marking, used for interactive design of target tracking and motion intention detection of objects, characterized in that: include: The color-based target tracking method is used to optimize the front settings; The motion intention enhancement detection method based on average gray value is used for post-optimization parameter adjustment; Whether the system is abnormal is to judge whether the system can perform the operation correctly; the threshold refers to setting relevant parameters and threshold size to detect markers; the baseline and optimal solution are action requests to be executed by the system; the image quality comparison is to calculate the structural clarity, signal-to-noise ratio and equivalent visual number of each frame of the image and other objective data to judge the accuracy of system recognition; when the system is judged to be abnormal, it is necessary to return to the first step for recalculation; when the tracking result before optimization is used as the baseline to assume the result of the system; when the result after optimization is higher than the result before optimization, the correctness of the assumption is proved to verify the feasibility and correctness of the system; when the action is executed, the system will give feedback to it to detect the object and the object's movement intention.

2. A method and system for target tracking and motion intention detection based on color marking, characterized in that: The specific core features are: using color markers as visual references to achieve real-time tracking of target objects, using the interpretable model CAM to select colors and spatial settings as the basis for color selection; The camera captures the scene and uses image processing algorithms to identify and extract color markers; motion detection algorithms are used to analyze the object's trajectory and speed to predict its movement intention; The core functions of the system include target recognition, motion analysis and intention prediction; in terms of user interaction, system settings can be adjusted or specific targets can be selected for tracking through touch screen or gesture control; the system can provide real-time feedback to users, such as displaying the current position and motion status of the target through a graphical interface to help users better understand and control the movement of the target; in terms of software, deep learning algorithms are integrated to improve recognition accuracy and real-time response capabilities; considering the user experience, the interface design should be simple and intuitive so that users can easily operate and adjust; the system can also explore the application of this technology in different fields, such as smart home, industrial automation or virtual reality; through continuous optimization of algorithms and user interaction design, this target tracking and motion intention detection method is expected to achieve wider application and improve the level of intelligence of human-computer interaction.

3. According to the method and system for target tracking and motion intention detection based on color marking as described in claim 1, wherein the determination involves the following steps: Step 1: Capture the scene image through the camera and identify the target marked with a specific color. Step 2: The system analyzes the target's motion trajectory and uses a prediction algorithm to determine its possible future motion direction and behavior. This process can be dynamically updated based on historical motion data and current speed, thereby achieving accurate judgment of the target's motion intention. Step 3: The system may combine other sensor data (such as accelerometer) to enhance the accuracy of judgment. Step 4: These judgment results can be fed back to the user in real time through the user interface, providing intuitive operation guidance and decision support.

4. According to a color-marking-based target tracking and motion intention detection method and system as described in claim 1, whether the system is abnormal further includes judging the parameter settings and system debugging of the system to execute according to the steps described in the system.

5. According to a method and system for target tracking and motion intention detection based on color marking as described in claim 2, the color marking is selected based on: the selected color has good contrast and recognition to ensure visibility under different background and lighting conditions.

6. The method and system for color-marked target tracking and motion intention detection according to claim 2, wherein the image processing algorithm includes color space conversion, edge detection and morphological processing to improve the extraction accuracy and reliability of color marks.

7. According to the color-marked target tracking and motion intention detection method and system described in claim 2, the motion detection algorithm adopts multiple feature fusion technologies, including but not limited to optical flow method, background subtraction and motion model, to improve the accuracy of target motion trajectory analysis.

8. The color-marked target tracking and motion intention detection method and system according to claim 2, wherein the system can provide real-time feedback to the user, and the real-time feedback is performed through sound, vibration or visual graphics to enhance the user's interactive experience.

9. The method and system for color-marked target tracking and motion intention detection according to claim 2, wherein the user interaction mode also includes voice command recognition to provide more convenient operation options. The system has self-learning capabilities and autonomously optimizes algorithm performance and user interface by continuously accumulating user interaction data and environmental change information.

10. The method and system for color-coded target tracking and motion intention detection according to claim 2, wherein the system is compatible with multiple sensor inputs, such as depth cameras, laser radars, etc., to enhance the accuracy of target detection and tracking. The algorithm can be adapted to different application scenarios, such as indoor and outdoor environments, dynamic and static scenes, to expand The application scope of the system. If you need to adapt to special scenarios, you need to set the system parameters. Just set the corresponding parameters to monitor it.

Images