Image fusion method based on plant community behaviors

Abstract

An image fusion method based on plant community behaviors optimizes image fusion operation by simulating behaviors of sowing, growing, flowering and results of a plant community, and comprises the following steps: step 1, registering and preprocessing an image to be fused by the plant community; step 2, performing sowing operation on the plant community in the to-be-fused image, and calculating image fusion parameters of plant individuals; step 3, carrying out growth operation on the plant community in the to-be-fused image and carrying out random search on image fusion parameters; 4, performing flowering operation on the plant community in the to-be-fused image, and randomly selecting image fusion parameters of neighbor plant individuals for combination; 5, performing result operation on the plant community in the to-be-fused image and mutually learning image fusion parameter information; and step 6, outputting an optimal fusion image by the plant community and ending the algorithm.

Description

technical field [0001] The invention belongs to the field of computer images and artificial intelligence, and particularly relates to an image fusion method based on plant community behavior. Background technique [0002] Image fusion refers to combining two or more source images obtained in different ways, or different source images obtained in the same way, using a fusion algorithm to form a new image, and comprehensively using the complementary advantages of different source images to obtain what a single image does not have. Information. Image fusion needs to extract the effective information of the source image, combine the respective advantages of different images, and improve the interpretation accuracy and recognition performance of the image through fusion technology. [0003] Image fusion technology first appeared in remote sensing image processing. In 1979, Daily et al. applied the fusion of Landsat-Mss images and radar images for the first time in geological in...

AI Technical Summary

Problems solved by technology

Traditional image fusion algorithms mainly rely on the information of the pixels themselves for fusion, but the information provided by the pixels themselves often does not consider the influence of the context, resulting in low fusion accuracy

When the image is disturbed by noise, or image sampling is inaccurate, or other factors interfere, the pixel information of the image is easy to change, resulting in poor accuracy of image fusion

[0014] 2. Algorithm efficiency is low

However, this type of image fusion method is easy to cause spectral distortion, and is usually used in occasions that do not require high image analysis accuracy.

Artificial intelligence algorithms often need to design loop nesting in the design process, and multiple loop nesting will lead to poor time performance and space performance of the algorithm

If the parameter design is not good, it is easy to affect the efficiency of the image fusion algorithm

[0015] 3. Low anti-interference ability

In practical applications, it is difficult to measure the spatial and temporal differences of image fusion, and there is no general error measurement standard at home and abroad

Image errors caused by spatio-temporal factors are difficult to eliminate directly using fusion algorithms

Although the traditional image fusion algorithm is easy to operate, it is easily disturbed by noise, and the operator is sensitive to noise

The Laplacian operator is more sensitive to noise interference, and even amplifies the adverse effects of noise on the image

[0016] 4. Poor scalability

However, when traditional image fusion methods ensure the resolution of multispectral images, it is difficult to ensure that the spectral characteristics of the images do not change.

In image fusion, how to maintain the spectral characteristics of the source image and at the same time improve the spatial resolution of the fused image is a problem that is difficult to balance

Images