A Method of Satellite Remote Sensing Image Classification Based on Rule Mining
A satellite remote sensing and classification method technology, applied in the field of aerospace remote sensing, can solve problems such as poor universality, unfavorable practical application, satellite remote sensing image analysis, etc. Effect
Active Publication Date: 2017-11-07
XIAN INSTITUE OF SPACE RADIO TECH
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] Although these research works have achieved certain results, these methods do not fully analyze the characteristics of satellite remote sensing images, resulting in complex classifier design, which is not conducive to practical applications; in addition, most of the current satellite remote sensing image classification methods use The features correspond to specific remote sensor types, and the universality is not good
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
[0030] The specific implementation of the present invention will be described below in conjunction with the accompanying drawings and specific implementation examples:
[0031] The images of the self-built satellite object image database of the present invention all come from the remote sensing satellite image database. The images are all in JPEG format with a gray scale of 256. The content of the images includes eight types of urban areas, sea ice, wheat fields, plains, sand seas, mountains, deserts, and tropical rainforests. The number of images contained in each type is different, and the size of the satellite ground object image library is 8 types of ground object images. The texture in some images has consistency in direction, and the structure distribution is uniform; the texture in some images does not have consistency in direction, and the structure distribution is irregular. Because it is a natural image, the texture information contained is more abundant.
[0032]...
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
A satellite remote sensing image classification method based on rule mining. The invention firstly performs multi-channel wavelet transformation on each satellite remote sensing image, and extracts the mean value and variance of the modulus of the wavelet transform coefficients of each channel of the image to form texture features. All satellite remote sensing The texture features of the image constitute the texture feature set; then the class association rule mining is carried out on the texture feature set, and the relationship between the feature set and the category set is directly mined, so as to select the most effective feature result set; finally, the decision tree is used to classify. The method of the invention is simple, can greatly reduce the complexity of the feature set of the remote sensing image on the satellite, and at the same time has a very high classification accuracy, can be applied to the rapid processing and distribution of the remote sensing image database on the satellite, and is a highly universal engineering method .
Description
technical field [0001] The invention relates to a method for classifying satellite remote sensing images, in particular to a method for classifying satellite remote sensing images based on rule mining, belonging to the field of aerospace remote sensing. Background technique [0002] With the increasing number and types of spaceborne remote sensors of earth observation satellites, especially high-resolution earth observation satellites, the resolution and data volume of acquired images are also developing rapidly, but the data urgently needs real-time and quasi-real-time transmission. , so the contradiction between satellite remote sensing image acquisition and data transmission is increasing day by day, and the existing data transmission system can no longer meet the requirements of rapid processing and distribution of satellite data in the future. If a satellite remote sensing image processing system with specific functions is established on the satellite, through prelimina...
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): G06K9/62
CPCG06F18/24
Inventor 张守娟马晓东陈瑞孙文方郑小松
Owner XIAN INSTITUE OF SPACE RADIO TECH