Ship target comprehensive identification method, medium and system
A recognition method and target recognition technology, which is applied in the comprehensive recognition method, medium and system fields of ship targets to achieve efficient recognition, improve recognition accuracy, and improve recognition accuracy
Pending Publication Date: 2022-05-17
江苏蓝视海洋科技有限公司
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Purpose of the invention: to provide an efficient and accurate ship target comprehensive recognition system, medium and method for the deficiency of the existing ship target single feature recognition technology
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0024] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.
[0025] Such as figure 1 As shown, the comprehensive identification method of ship target includes the following steps: preprocessing the infrared video of the ship, preprocessing the side scan sonar data set, preprocessing the ship radiation noise data; extracting ship The characteristics of each data; build a learning model adapted to the video data set for the infrared video data of the ship, build a GAN generation confrontation network learning model for the side-scan sonar data set, and build a semi-supervised learning model adapted to the audio data for the radiation noise data of the ship; Three sets of classification results are used to obtain the recognition accuracy rate, and the final recognition accuracy rate is obtained based on the GA merging network model and weighted voting decision fusion method.
[0026] The recognition resul...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a ship target comprehensive identification method. The method comprises the following steps: respectively acquiring and preprocessing ship infrared video data, side-scan sonar image data and ship radiation noise data; heterogeneous learning models suitable for data are constructed for the ship infrared video data set, the side-scan sonar image data set and the ship radiation noise data set respectively and are trained, and corresponding classification results of three groups of different networks are obtained; network parameters are synchronously updated according to the change of test data, the model matching degree is improved, and the accuracy of each recognition model is improved; and putting the identification results of the three groups of heterogeneous data into an updated voting decision theory algorithm, and fusing the identification results of the three groups of ship target data to obtain a final ship target comprehensive identification result. According to the method, the influence of low accuracy of single target feature recognition is avoided, the recognition accuracy can be effectively improved by combining the training model and the decision algorithm, the sea surface ship target is efficiently recognized, and the method has feasible practical value and application prospect.
Description
technical field [0001] The invention relates to the field of ship target recognition, in particular to a ship target comprehensive recognition method, medium and system. Background technique [0002] In recent years, with the rapid development of the national marine economy, the demand for the recognition of sea surface ship targets is increasing rapidly. Using video recognition technology can effectively identify the type and hull number of the target ship on the sea surface; using infrared video recognition at the end can identify the heat information of the ship's engine, thereby distinguishing the ship; using image sonar technology can identify the movement of the ship's underwater target, and monitor The safety status of underwater sports; the type and hull number of ships in the water can be distinguished by using radiation noise recognition technology, and the data of different signal sources can be recognized by using different recognition methods, but the recognitio...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06V20/10G06K9/62G06V10/774G06V10/80G06V10/764
CPCG06F18/241G06F18/25G06F18/214
Inventor 张友文葛慧林王彪刘鹏
Owner 江苏蓝视海洋科技有限公司