11results about How to "Improve target positioning accuracy" patented technology

The invention discloses a traffic sign detection method and device based on YOLOv5 and a medium, belongs to the technical field of traffic sign detection, and aims to solve the technical problem of how to improve the small target detection precision, reduce the omission ratio and further improve the detection performance of traffic signs. According to the technical scheme, a prior anchor point frame is optimized by using a genetic algorithm and a K-means algorithm, and the detection accuracy of a small target is enhanced; a Bi-FPN structure is introduced to realize feature fusion of semantic and position information, and the representation capability of the network is improved; by integrating an attention mechanism GAM, the anti-interference performance and the feature extraction capability of the network in a complex environment are enhanced; sIOUloss is adopted as a frame regression loss function, so that the positioning accuracy is improved.

The invention discloses an autonomous acousto-optic fusion target detection method and system for an amphibious unmanned platform, relates to the technical field of underwater target detection, and aims to solve the problem that a method for detecting a complex underwater environment target by combining an acoustic sensor and an optical sensor is lacked in the prior art. Comprising the following steps: S1, synchronously acquiring underwater environment information through an optical camera and a foresight sonar of the amphibious unmanned platform; s2, preprocessing the image data and the acoustic data respectively, and then performing segmentation and target identification on the preprocessed image data and acoustic data; s3, respectively converting target identification results under the optical camera coordinate system and the foresight sonar coordinate system to a joint coordinate system; and S4, establishing a support degree matrix between the target identification results of the optical camera and the foresight sonar, rejecting outliers with low support degree according to the obtained comprehensive support degree matrix of each data, and performing optimal weighted fusion on measured values with high support degree to obtain an acousto-optic combined underwater target positioning result.

The invention discloses a robot motion control system and control method based on machine vision, relates to the technical field of machine vision, and provides effective data input for an error calculation and correction module by extracting a local feature point set Fp during operation of the system. And the error calculation and correction module calculates an image translation error Et and a rotation error Er based on the feature point set Fp, obtains a translation vector T and a rotation error R by correcting the image position, generates a complete panoramic image P, and provides reliable visual information support for subsequent target positioning and path planning. The motion control and target positioning module is combined with a panoramic image P and a positioning algorithm, so that the target positioning precision is improved, the path planning is optimized, the problems of image dislocation, positioning errors and path instability in a traditional method are solved, and the adaptive capacity and task execution efficiency of the robot in a dynamic environment are greatly improved.

This application provides a vehicle control method for a two-wheeled vehicle and a two-wheeled vehicle. The method includes: during the operation of the two-wheeled vehicle, acquiring current attitude data collected by an attitude sensor and current perception data collected by a sensing device, wherein the acquisition time of the current perception data matches the acquisition time of the current attitude data, and the current perception data is perception data in the device coordinate system of the sensing device; correcting a preset mapping relationship between the device coordinate system and the world coordinate system based on the current attitude data to obtain a current mapping relationship; performing coordinate transformation on the current perception data using the current mapping relationship; performing target identification on the coordinate-transformed current perception data to obtain a set of perceived targets; and controlling the two-wheeled vehicle based on the set of perceived targets. This application solves the problem of low target positioning accuracy caused by unstable sensor coordinate systems in related technologies, thereby improving target positioning accuracy.

The invention discloses an individual multifunctional reconnaissance system, which is characterized by comprising an infrared detection module, a laser ranging module, a visible light module, a positioning module, an attitude measurement module and a comprehensive information processing module, the infrared detection module, the laser ranging module, the visible light module, the positioning module and the attitude measurement module are all connected with the comprehensive information processing module; the comprehensive information processing module is used for receiving detection data of each module, carrying out target calculation and estimating a target motion state by adopting Kalman filtering; the method has the advantages that the target positioning method is simple in calculation and high in positioning precision.

According to the ornithopter reconnaissance system disclosed by the invention, a lightweight dual-light pod is adopted, so that the reconnaissance capability is enhanced compared with a traditional pod while the load weight of a ornithopter flying robot is reduced; the three-axis pod adapting to the motion form of the flapping-wing flying robot is adopted, so that jitter of a reconnaissance video during flying of the flapping-wing flying robot is greatly improved; a pod reconnaissance system combining long and short focuses is adopted, so that the accurate searching capability is improved; algorithms of airborne target detection, recognition, tracking and the like are integrated, and the problem that a small-sized flapping-wing flying robot cannot carry out airborne reconnaissance video real-time processing is solved; by adopting the cluster reconnaissance method, the target positioning precision is improved, and accurate target positioning and cooperative reconnaissance capability can be realized.

The invention discloses a drop-out fuse state monitoring method based on multi-size features and deep learning, and the method comprises the steps: S1, forming a data set for obtained drop-out fuse pictures, and dividing the data set into a training set and a test set; s2, adding an improved receptive field block and a coordinate attention module to a YOLOx backbone network, adding an adaptive spatial feature fusion module to PANet, carrying out secondary fusion on features of different scales, then introducing a loss function of weighting loss and positioning loss fusion, and finally carrying out lightweight improvement, constructing a state monitoring model, and carrying out state monitoring. Performing training verification on the constructed state monitoring model through the training set and the test set; s3, identifying a drop-out fuse picture acquired in real time by adopting the trained and verified state monitoring model; according to the image data collected by the application, the construction of a special database is realized, the YOLOx is improved and lightweight operation is carried out, and the complexity of the model is reduced, so that the method is suitable for an embedded platform of an electric unmanned aerial vehicle, and the detection effect of the drop-out fuse is improved.

The application discloses a three-dimensional target tracking method, system, device and medium based on space-time enhancement, and relates to the technical field of automatic driving. The method comprises the following steps: inputting three-dimensional point cloud sequence data into a trained SMTrack network for processing to obtain a final bounding box prediction result. The SMTrack network comprises a target-specific encoder, an STFM module and an STT module connected in sequence. The target-specific encoder is used for extracting target-specific features from the point cloud sequence. The STFM module is used for modeling appearance information and motion information in stages according to the target-specific features to generate a preliminary bounding box prediction result. The STT module is used for optimizing the preliminary bounding box prediction result. The application adopts a twin structure, extracts features from historical frame and current frame sequences, and introduces part perception motion modeling and coarse-to-fine bounding box regression mechanisms, so that the target positioning accuracy can be greatly improved.

The invention relates to the field of underwater detection devices, and discloses an underwater frogman detection sonar device based on a beam forming technology, which comprises a main console and a plurality of detection sub-devices, and is characterized in that the main console is in two-way communication connection with each detection sub-device; an identification module of the static identification total unit is used for acquiring the received signal and judging whether the signal-to-noise ratio DT of the received signal is greater than a preset target appearance threshold value, if the judgment result is false, judging the received signal as a low-activity signal and re-acquiring the received signal, and if the judgment result is true, judging that the underwater target to be detected appears, and otherwise, judging that the underwater target to be detected appears. The detection sub-device receives a signal, judges the received signal as a non-low-activity signal, and sends the signal to the control module of the detection sub-device, the control module of the detection sub-device performs information transmission with the communication module of the main console through the communication module, and the main console controls the dynamic identification total unit and the transmitting transducer to perform a dynamic identification process. According to the device, active and passive detection processes are combined, so that the underwater frogman detection process is more accurately carried out.

The invention provides an external radiation source detection direct wave path error correction method, device and equipment, a medium and a program product, and the method comprises the steps: S1, constructing a relation between a known position information target and an unknown position information target according to a time difference measurement result; and S2, detecting a target with known position information, and positioning the target with the unknown position information by using the relationship between the target with the known position information and the target with the unknown position information. According to the method, correction of the direct wave path error is completed based on prior information, the problem that the positioning error is large when a non-intervisible external radiation source is used for external radiation source detection is solved, and the method enables the performance of a passive detection system based on the external radiation source to be more stable and the target positioning precision to be higher.

The invention provides a target positioning precision optimization method and device based on road side fusion perception. The method comprises the following steps: acquiring latitude and longitude information of a centroid point of a target object through a millimeter-wave radar on a road side, and acquiring pixel position information of a key point of the target object through a camera on the road side; determining sensing positioning data of the target object based on the latitude and longitude information of the centroid point of the target object and the pixel position information of the key point of the target object; determining the target positioning precision based on the sensing positioning data and the real positioning data of the target object; and optimizing the target positioning precision by adjusting the plane position of the centroid point of the target object and / or the latitude and longitude information of the centroid point of the target object until the target positioning precision reaches a preset precision requirement. According to the target positioning precision optimization method and device based on the roadside fusion perception, the material center position of the target object is optimized according to the differential expression of the perception data and the truth value data, and the target object positioning precision of the roadside fusion perception is improved.