Finger vein recognition method based on fusion of local features and global features

Abstract

The invention discloses a finger vein recognition method fusing local features and global features. At present, a number of vein recognition methods adopt the local features of a vein image, so that the recognition precision of the vein recognition methods is greatly affected by the quality of the image; the phenomena of rejection and false recognition are liable to appear. The finger vein recognition method provided by the invention comprises the following steps: firstly, performing pretreatment operations such as finger area extraction of a read-in finger vein image, binarization and the like; then, according to the point set of extracted detail features, realizing the matching of the local features within a certain angle and a certain radius by virtue of a flexible matching-based local feature recognition module; using a global feature recognition module for vein image recognition to realize the matching of the global features as the global feature recognition module is used for analyzing bidirectional two-dimensional principal components and can better display a two-dimensional image data set on the whole; finally, designing weights according to the correct recognition rates of the two recognition methods, performing decision-level fusion to the results of two classifiers, and taking the fused result as a final recognition result. The method is applied to finger vein recognition.

Description

Technical field:

[0001] The invention relates to a finger vein recognition method which combines local features and global features. Background technique:

[0002] At present, many vein recognition methods use local features of vein images, which are mainly used to describe the detailed information of finger vein patterns. In the prior art, the matching recognition operation is completed by extracting the minutiae features on the vein pattern, and then comparing the minutiae features of the vein image to be matched with the template image. This method can obtain a higher recognition rate, and the recognition speed Faster, but it is greatly affected by image quality. For vein images with poor quality, it is easy to miss feature points or extract false feature points, and it is difficult to accurately extract detail point features, resulting in rejection in the matching decision process. , The phenomenon of misunderstanding. In addition to the recognition method based on loc...

Examples

Embodiment 1

[0023] A finger vein recognition method that integrates local features and global features. Firstly, preprocessing operations such as finger area extraction and binarization are performed on the read finger vein images; The feature recognition module realizes the matching of local features within a certain angle and radius; the global feature recognition module for two-way two-dimensional principal component analysis can better display the two-dimensional image data set as a whole, and use it for vein Image recognition realizes the matching of global features; finally, weights are designed according to the correct recognition rates of the two recognition methods, and the results of the two classifiers are fused at the decision level, and the fused result is taken as the final recognition result.

Embodiment 2

[0025] According to the finger vein recognition method for fusing local features and global features described in Embodiment 1, the flexible matching local feature recognition module,

[0026] (1) First, extract the finger area, binarize, and thin the original finger vein image, and then extract the feature points of the thinned image, that is, the endpoint and the intersection point;

[0027] (2) Read in template image feature points and feature points of the image to be matched ,judge with Is it satisfied , if not established, repeat this step and read in another pair of feature points, otherwise turn to step (3), until all the details point pairs are compared, go to the last step;

[0028] (3) Accumulate the number of similar feature points;

[0029] (4) According to the following similarity calculation formula, the matching similarity between the template image and the feature point set of the image to be matched is obtained, and compared with the qualified thres...

Embodiment 3

[0033] According to the finger vein recognition method of fusing local features and global features described in Embodiment 1 or 2, the decision-making level fusion, after obtaining the recognition method based on flexible matching and the finger vein recognition method based on two-dimensional two-dimensional principal component analysis After the correct recognition rate and the recognition result, the weights are designed according to the correct recognition rate of the two recognition methods, so as to determine the proportion of the recognition results of the two recognition methods in the final fusion result; after obtaining the global feature method and Correct recognition rate of local feature method with and the recognition results of these two recognition systems with Finally, the final recognition result is obtained by linear fitting; among them, with The value of is 1 or 0, 1 indicates that the system verification is successful, and 0 indicates that the v...

Technical field:

[0001] The invention relates to a finger vein recognition method which combines local features and global features. Background technique:

[0002] At present, many vein recognition methods use local features of vein images, which are mainly used to describe the detailed information of finger vein patterns. In the prior art, the matching recognition operation is completed by extracting the minutiae features on the vein pattern, and then comparing the minutiae features of the vein image to be matched with the template image. This method can obtain a higher recognition rate, and the recognition speed Faster, but it is greatly affected by image quality. For vein images with poor quality, it is easy to miss feature points or extract false feature points, and it is difficult to accurately extract detail point features, resulting in rejection in the matching decision process. , The phenomenon of misunderstanding. In addition to the recognition method based on loc...

Examples

Embodiment 1

[0023] A finger vein recognition method that integrates local features and global features. Firstly, preprocessing operations such as finger area extraction and binarization are performed on the read finger vein images; The feature recognition module realizes the matching of local features within a certain angle and radius; the global feature recognition module for two-way two-dimensional principal component analysis can better display the two-dimensional image data set as a whole, and use it for vein Image recognition realizes the matching of global features; finally, weights are designed according to the correct recognition rates of the two recognition methods, and the results of the two classifiers are fused at the decision level, and the fused result is taken as the final recognition result.

Embodiment 2

[0025] According to the finger vein recognition method for fusing local features and global features described in Embodiment 1, the flexible matching local feature recognition module,

[0026] (1) First, extract the finger area, binarize, and thin the original finger vein image, and then extract the feature points of the thinned image, that is, the endpoint and the intersection point;

[0027] (2) Read in template image feature points and feature points of the image to be matched ,judge with Is it satisfied , if not established, repeat this step and read in another pair of feature points, otherwise turn to step (3), until all the details point pairs are compared, go to the last step;

[0028] (3) Accumulate the number of similar feature points;

[0029] (4) According to the following similarity calculation formula, the matching similarity between the template image and the feature point set of the image to be matched is obtained, and compared with the qualified thres...

Embodiment 3

[0033] According to the finger vein recognition method of fusing local features and global features described in Embodiment 1 or 2, the decision-making level fusion, after obtaining the recognition method based on flexible matching and the finger vein recognition method based on two-dimensional two-dimensional principal component analysis After the correct recognition rate and the recognition result, the weights are designed according to the correct recognition rate of the two recognition methods, so as to determine the proportion of the recognition results of the two recognition methods in the final fusion result; after obtaining the global feature method and Correct recognition rate of local feature method with and the recognition results of these two recognition systems with Finally, the final recognition result is obtained by linear fitting; among them, with The value of is 1 or 0, 1 indicates that the system verification is successful, and 0 indicates that the v...

Images