Hyperspectral image classification method, storage medium and computer equipment
A hyperspectral image and classification method technology, applied in the field of storage media and computer equipment, and hyperspectral image classification methods, can solve problems such as inability to classify hyperspectral images, and achieve the effects of improving discrimination ability, accurate classification accuracy, and reducing noise
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0035] like figure 1 As shown, a hyperspectral image classification method, the image classification method performs the following steps:
[0036] Step S1: First, for the hyperspectral data, in order to make the input data smoother, it is normalized and preprocessed so that the value range of the hyperspectral data set is between 0 and 1.
[0037] The normalized preprocessing formula described therein is as follows:
[0038] x ij =x ij * / max(X) (1)
[0039] where x ij * Represents a piece of data in the hyperspectral data set, and max() represents the largest data in the hyperspectral data set.
[0040] Step S2: setting several different neighborhood window scales; the neighborhood window scales are 3, 5, 7, 9, 2n+1 respectively. where denotes the nth neighborhood window.
[0041] Step S3: Pass the original hyperspectral data through a neighborhood window g of a certain scale, and use a weighted mean filter to perform noise reduction and edge extraction features on t...
Embodiment 2
[0068] Based on the hyperspectral image classification method described in Embodiment 1, this embodiment also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the processor executes Follow the steps below:
[0069] S1: Perform normalized preprocessing on the hyperspectral data, so that the value range of the hyperspectral data set is between 0 and 1;
[0070] S2: Set the scale of several different neighborhood windows
[0071] S3: Pass the original hyperspectral data through a neighborhood window g of a certain scale, and use a weighted mean filter to perform noise reduction and edge extraction features on the image;
[0072] S4: pass the original hyperspectral data through a neighborhood window g of a certain scale, obtain the attribute profile of the hyperspectral image, and obtain the extended multi-attribute profile of the hyperspectral image, and then use a weighted mean filter to reduce noi...
Embodiment 3
[0076] A computer device, comprising a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:
[0077] S1: Perform normalized preprocessing on the hyperspectral data, so that the value range of the hyperspectral data set is between 0 and 1;
[0078] S2: Set the scale of several different neighborhood windows
[0079] S3: Pass the original hyperspectral data through a neighborhood window g of a certain scale, and use a weighted mean filter to perform noise reduction and edge extraction features on the image;
[0080] S4: pass the original hyperspectral data through a neighborhood window g of a certain scale, obtain the attribute profile of the hyperspectral image, and obtain the extended multi-attribute profile of the hyperspectral image, and then use a weighted mean filter to reduce noise and smooth the hyperspectral Homogeneous regions in the image, using weighted e...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More