High-resolution image matching method and system

1. A high-resolution image matching method, comprising the following steps:

performing regional fidelity down-sampling on an initial high-resolution image to obtain a multi-level low-resolution image;

performing local matching on the obtained multi-level low-resolution images using a method with global probes to obtain a matching result of the low-resolution image; and

performing reverse refinement on the obtained matching result of the low-resolution image using overall consistency of the image matching, to obtain the matching results of the high-resolution images at all levels until the matching results of the initial resolution images are obtained, wherein,

in the obtained matching result of the lower resolution image, performing pairing processing on each matched pixel to find their respective corresponding pixel regions in the upper level of resolution image;

performing a matching calculation on the pixels within the region for the corresponding pixel region in the obtained upper level image of higher resolution, to obtain the optimal pixel matching relationship within the pixel region;

determining the obtained optimal pixel matching relationship within the pixel region as the matching relationship between the current resolution images; and

cyclically performing the operation steps of matching calculation of the optimal pixel matching relationship within the pixel region and determination of the matching relationship between the current resolution images, until the matching result of the highest resolution image is obtained and taken as the final initial resolution image matching result.

2. The high-resolution image matching method according to claim 1 , wherein, the step of performing regional fidelity down-sampling on an initial high-resolution image to obtain a multi-level low-resolution image specifically comprises the following steps:

initializing the value of the number of pixels R, wherein the number of pixels R is taken as a sampling stopping condition;

taking the obtained value of the number of pixels R as the pixel value at the corresponding position of the next level of resolution image; and

performing the above down-sampling process in sequence, until the obtained image resolution satisfies the sampling stopping condition, and stopping the sampling process, wherein the finally obtained image is the lowest resolution image.

3. The high-resolution image matching method according to claim 2 , wherein, in the initial high-resolution image process, the information of four pixels adjacent to each other above and below and on the left and right is synthesized to obtain a value, and each obtained value contains the information of the four adjacent images.

4. The high-resolution image matching method according to claim 2 , wherein, the step of performing local matching on the obtained multi-level low-resolution image using a method with global probes to obtain a matching result of the low-resolution image specifically comprises the following steps:

initializing the number, direction and length of global probes;

performing a pixel-by-pixel matching on the obtained lowest resolution image, and calculating the matching cost corresponding to each matching result;

calculating the magnitude relationship between the pixel value and the center pixel value of each position of the lowest resolution image on the probe according to the determined number, direction and length of probes, and representing in a binary value;

calculating the distance between the probes corresponding to all the pixels to be matched;

performing weighted summation on each pixel to be matched based on the calculated matching cost corresponding to each matching result and the obtained distance between corresponding probes, wherein the obtained weighted summation result is taken as the final matching cost of each pixel to be matched; and

selecting the candidate pixel with the lowest matching cost as the final matching result of the low resolution image.