A fast mosaic method for UAV images based on machine learning and feature point recognition
A fast stitching and machine learning technology, applied in the field of panoramic image stitching, can solve problems such as unsatisfactory stitching effect, low stitching efficiency, and many redundant messages, and achieve the effects of improved accuracy, fast operation, and improved stitching efficiency
Active Publication Date: 2022-04-12
CHINA INST OF WATER RESOURCES & HYDROPOWER RES +1
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the present invention is to solve the problems of unsatisfactory splicing effect, many redundant messages, low splicing efficiency and long splicing time in the existing fast splicing method of UAV images, and proposes a method based on machine learning and feature point recognition Fast splicing method for UAV images
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
[0070] Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the implementations shown and described in the drawings are only exemplary, intended to explain the principle and spirit of the present invention, rather than limit the scope of the present invention.
[0071] The embodiment of the present invention provides a method for fast splicing of UAV images based on machine learning and feature point recognition, such as figure 1 As shown, the following steps S1-S11 are included:
[0072] S1. Obtain the image data collected by the UAV aerial photography in real time, randomly specify the reference image from the image data, and determine the 9 adjacent images of the reference image through the latitude and longitude information during the UAV aerial photography process.
[0073] S2. Use the improved SURF algorithm based on GPU parallel acceleration optimization to simultaneousl...
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 for fast splicing of unmanned aerial vehicle images based on machine learning and feature point recognition. First, a reference image is selected, and its surrounding 9 adjacent images are characterized using the improved SURF algorithm based on GPU parallel acceleration optimization. The extraction of points reduces the time it takes to extract feature points from the reference image multiple times, and reduces the cumulative error of image multiplication, and then uses the description feature vector calculated by machine learning to improve the description feature vector of the SURF algorithm, which greatly improves the matching of feature points Accuracy for image registration. Based on the SURF algorithm + machine learning + GPU + PROSAC algorithm + image block stitching technology, the present invention constructs a fast and efficient processing method for UAV remote sensing images, which runs faster than the traditional SURF algorithm and greatly improves the accuracy , and its real-time performance is more prominent.
Description
technical field [0001] The invention belongs to the technical field of panorama image stitching, and in particular relates to the design of a method for fast stitching of drone images based on machine learning and feature point recognition. Background technique [0002] Panoramic image stitching has extensive research literature and several commercial applications in areas such as photogrammetry, computer vision, image processing, and computer graphics. With the application of drones in military reconnaissance, disaster relief, forest fire monitoring, remote sensing and telemetry, and other fields, its unique characteristics have attracted more and more domestic and foreign experts and scholars to invest in the key technology of drones in the research. UAV image stitching technology is the use of image stitching technology to stitch the aerial images of UAVs to form an intuitive and easy-to-understand image of a large scene. Due to the large amount of data of aerial images...
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 Patents(China)
IPC IPC(8): G06T3/40G06N3/08G06N3/04G06K9/62G06V10/46G06V20/17G06V10/764
CPCG06T3/4038G06N3/082G06N3/08G06V10/462G06N3/045G06F18/24Y02T10/40
Inventor 雷添杰邓安军张春再李爱丽胡海华徐瑞瑞王党伟王嘉宝宫阿都
Owner CHINA INST OF WATER RESOURCES & HYDROPOWER RES