Forest Edge Detection Based On High Spatial Resolution Remote Sensing Imagery And Wavelet Analysis

As computers and the rapid development of information technology have been widely used in contour, feature extraction, texture analysis and other fields, edge detection becomes an important image processing technology. .The wavelet analysis used in signal processing and image processing is a research hotspot in recent years because of its unique advantages in multi-resolution analysis of the characteristics. This paper focuses on the applications of wavelet analysis in the forest edge information extraction, including trees, shrubs and bamboo plant identification and edge detection of large area forest.Firstly, this paper briefly introduces the basic theory of wavelet analysis and its wide range of applications, and analyzes the advantages and disadvantages of various edge detection extraction methods based on remote sensing images. Secondly, different type of forests'edge are detected using wavelet transform modulus maxima method according to multi-resolution wavelet analysis features and the different vegetation's texture with different frequencies based on high resolution resolution image, such as WorldView satellite images, aerial drones, etc.The main bodies of this paper are as followed. High-resolution remote sensing images(WorldView satellite images, aerial drones, etc.) with forest are transformed into vertical and horizontal detail components using wavelet transform, and then modulus maximum and the phase obtained angle are calculated using these two detail components. Those modulus maximum less than threshold were removed as noise points. The forest edge was detected using the remaining points.In this paper, the crown edges of single tree, shrubs, bamboo were detected using the above method. Comparison between WorldView satellite images and aerial drones showed that accuracy for WorldView satellite images was higher than that for aerial drones because of higher resolution and complex information of aerial drones. The edge of lager area bamboo forest can be effectively detected based on WorldView satellite. Finally, results were verified using field inventory data.Most home and abroad studies on edge detection using wavelet analysis are based on regular images and less based on remote sensing images. Studies on highly irregular forest edge detection with complex information are rare. Through a large number of experiments, this paper proves that wavelet analysis is a feasible method to detect forest edge using remote sensing images,...