Water area detection method and device based on visual monitoring of empty base platform
A detection method and water area technology, applied in the field of image processing, can solve problems such as difficulty in monitoring water area, and achieve the effects of convenient local debugging, accurate segmentation and strong robustness
Active Publication Date: 2019-01-18
BEIHANG UNIV
View PDF6 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The present invention introduces hyperspectral pictures to monitor changes in water area, specifically a water area detection method and device based on space-based platform visual monitoring, to solve the existing difficult problem of water area monitoring
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
[0042] The present invention will be further described in detail with reference to the accompanying drawings and embodiments.
[0043] The water area detection method and device based on space-based platform visual monitoring provided by the present invention first adopts a method based on CNN (Convolution neural network) and atrous convolution to extract edge information in the spatial information of the picture. Secondly, according to the division of edge information, all the pixels of the original image are grouped. In order to obtain a stable classification result, the expectation of the spectral information of each group of pixels is calculated as the final sample vector for classification. Use the linear SVM classifier again to train the model, predict water areas and non-water areas, and output classification results. Finally, the real area of the water area is calculated according to the legend parameter gamma obtained from the camera parameters and the space-based ...
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 method and a device for detecting water area based on visual monitoring of an empty base platform, which relates to the technical field of image processing. The device comprises an image acquisition device, a CNN feature extractor, a branching device of a porous convolution network, a feature fusion device, a classifier and an area converter. Firstly, the convolution neural network extracts the high-level semantic information from the images of the water area to be monitored, obtains the output feature map, and inputs it to the porous convolution network. Then, according to the edge information, the hyperspectral information of all pixels in each region is averaged to obtain the desired characteristics of the region. The training model of linear SVM classifier isused to classify the sample vectors of desired features, and the classification results are output to obtain the water area S. The true area of the water area is calculated according to the legend parameter gamma and the water area S of the original map. The invention has strong robustness, can monitor the change of the real area of the water area part and predict the natural disaster.
Description
technical field [0001] The invention relates to the technical field of image processing, in particular to a water area detection method and device based on the visual monitoring of a space-based platform. Background technique [0002] The monitoring of water area refers to the monitoring and extraction of water areas based on aerial images combined with specific geographic information data. The monitoring of water area can measure the size of waterfalls, lakes, rivers, reservoirs and other water areas in real time, so as to monitor and predict natural disasters such as heavy rain, mudslides and drought, and help people take protective measures in time to avoid harm, which has very important practical significance. [0003] Using traditional methods to monitor water areas often consumes huge manpower and material resources. However, the monitoring method based on the space-based platform visual surveillance system has the characteristics of mobility, large amount of data and ...
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 Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06T7/11G06T7/62
CPCG06T7/11G06T7/62G06N3/045G06F18/2411G06F18/253
Inventor 曹先彬甄先通李岩沈佳怡
Owner BEIHANG UNIV