Palm vein recognition login method based on domestic CPU and operating system
Through the palm vein recognition and login method based on domestic CPU and operating system, using infrared cameras and advanced image processing algorithms, the security and accuracy of the existing login method are solved, and high security, high accuracy and non-contact Format authentication, suitable for a variety of occasions.
Patent Information
- Application Number
- CN202411269725.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-17
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing login methods such as passwords, PIN codes, fingerprints and facial recognition have problems such as low security, susceptibility to environmental impact and inconvenience in use, especially in the field of information and innovation, it is difficult to meet the needs of high security and accuracy.
Using the palm vein recognition and login method based on domestic CPU and operating system, the palm vein images are captured through infrared cameras, and denoising, contrast enhancement, segmentation and feature extraction are performed, combining Gaussian blur, median filtering, bilateral filtering, watershed algorithm and hog features Extraction, use nearest neighbor algorithm to perform feature matching verification to achieve identity verification.
It realizes high security, high accuracy and contactless authentication, which is suitable for many occasions, especially in areas with high security requirements, and has a wide range of applications.
Smart Images
Figure CN119150269A_ABST
Abstract
Description
Technical Field
[0001] The invention discloses a palm vein recognition login method based on a domestically produced CPU and an operating system, and relates to the technical field of image recognition. Background Art
[0002] With the country's strong support for information technology innovation, the industry has developed rapidly. A number of excellent information technology innovation software and hardware products have emerged in China, which can already meet the daily use of domestic governments, enterprises and institutions. More and more Web application software systems based on information technology innovation software and hardware have been put into use in many industries.
[0003] Common login methods usually rely on passwords, PIN codes, or biometric technologies such as fingerprints and facial recognition. These methods have some disadvantages: For example, passwords and PIN codes: users need to remember complex passwords or PINs, which are easy to forget or confuse, resulting in frequent reset operations. In addition, passwords and PIN codes are easy to guess or obtain through social engineering means, and their security is low; fingerprint recognition: although convenient, it is greatly affected by the surface state of the finger, such as stains, moisture or damage may cause recognition failure. In addition, fingerprint readers are usually exposed to the outside and are easily damaged by environmental factors. After long-term use, the recognition accuracy may decrease; facial recognition: affected by ambient light, user posture and facial occlusion (such as masks), it may affect the recognition efficiency and accuracy. Summary of the invention
[0004] In view of the problems of the prior art, the present invention provides a palm vein recognition login method based on a domestically produced CPU and an operating system, which has the characteristics of strong versatility and simple implementation, and has broad application prospects.
[0005] The specific scheme proposed by the present invention is:
[0006] The present invention provides a palm vein recognition login method based on a domestically produced CPU and an operating system, comprising:
[0007] Step 1: Based on domestic CPU and operating system, input palm vein data:
[0008] Step 11: Use an infrared camera to capture the reflection image of the palm vein under infrared light to obtain a palm vein image.
[0009] Step 2: Preprocess the palm vein image:
[0010] De-noising of palm vein images.
[0011] Perform contrast enhancement on palm vein images.
[0012] Adjust the size of the palm vein image,
[0013] Segment the palm vein image and perform rotation correction on the palm vein image.
[0014] The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent.
[0015] Extracting features of relevant vein patterns of palm vein images;
[0016] Step 3: Verify palm vein data: Perform feature matching verification on the pre-processed palm vein image and the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
[0017] Furthermore, in step 2 of the palm vein recognition login method based on a domestic CPU and operating system, when the palm vein image is denoised, Gaussian blur, median filtering, and bilateral filtering algorithms are used to remove noise.
[0018] When contrast enhancement is performed on palm vein images, the limited contrast adaptive histogram equalization method is used to increase the distinction between image regions and highlight the vein structure.
[0019] When resizing palm vein images, resize the palm vein images to a uniform size and resolution so that all images meet the same size standards.
[0020] When segmenting the palm vein image and performing rotation correction on the palm vein image, the palm vein area is separated from the background using the image segmentation method, and the palm vein image is angle-corrected to ensure consistency between images and facilitate subsequent feature matching.
[0021] Furthermore, when segmenting the palm vein image in step 2 of the palm vein recognition login method based on a domestic CPU and an operating system, it also includes: performing morphological processing on the segmented palm vein image, binarizing the segmented palm vein image and performing connectivity analysis on the segmented palm vein image, performing hog feature extraction on the segmented palm vein image, and constructing a feature vector.
[0022] Furthermore, in step 2 of the palm vein recognition login method based on a domestic CPU and an operating system, when performing morphological processing, a watershed algorithm is used to refine the segmentation of the vein pattern to avoid incorrect connections between veins; when performing binarization processing and connectivity analysis, the image after morphological processing is binarized by a preset threshold, the veins and the background are clearly distinguished, a binary image is obtained, the network structure belonging to the veins in the binary image is determined, and isolated pixels or small areas are removed to obtain a clean vein pattern; when performing hog feature extraction, the directional gradient histogram of the veins in the palm vein image is calculated to obtain the hog feature, and the shape, size and direction information of the veins are captured according to the hog feature.
[0023] Furthermore, in step 3 of the palm vein recognition login method based on a domestic CPU and operating system, a nearest neighbor algorithm is used to perform feature matching verification, and the palm vein image after preprocessing is classified by comparing the similarity with the image in the palm vein image database, and compared with a preset threshold to determine whether the palm vein image is successfully matched.
[0024] The present invention provides a palm vein recognition login device based on a domestically produced CPU and an operating system, comprising an input module, a collection module, a preprocessing module and a verification and recognition module.
[0025] The input module is based on domestic CPU and operating system, and inputs palm vein data:
[0026] The collection module uses an infrared camera to capture the reflection image of the palm vein under infrared light to obtain the palm vein image.
[0027] The preprocessing module preprocesses the palm vein image:
[0028] De-noising of palm vein images.
[0029] Perform contrast enhancement on palm vein images.
[0030] Adjust the size of the palm vein image,
[0031] Segment the palm vein image and perform rotation correction on the palm vein image.
[0032] The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent.
[0033] Extracting features of relevant vein patterns of palm vein images;
[0034] The verification and recognition module verifies the palm vein data: the pre-processed palm vein image is feature matched with the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
[0035] Furthermore, when the preprocessing module of the palm vein recognition and login device based on the domestic CPU and operating system performs denoising on the palm vein image, Gaussian blur, median filtering, and bilateral filtering algorithms are used to remove noise.
[0036] When contrast enhancement is performed on palm vein images, the limited contrast adaptive histogram equalization method is used to increase the distinction between image regions and highlight the vein structure.
[0037] When resizing palm vein images, resize the palm vein images to a uniform size and resolution so that all images meet the same size standards.
[0038] When segmenting the palm vein image and performing rotation correction on the palm vein image, the palm vein area is separated from the background using the image segmentation method, and the palm vein image is angle-corrected to ensure consistency between images and facilitate subsequent feature matching.
[0039] Furthermore, when the preprocessing module of the palm vein recognition login device based on the domestic CPU and operating system segments the palm vein image, it also includes: performing morphological processing on the segmented palm vein image, binarizing the segmented palm vein image and performing connectivity analysis, performing hog feature extraction on the segmented palm vein image, and constructing a feature vector.
[0040] Furthermore, when the preprocessing module of the palm vein recognition login device based on the domestic CPU and operating system performs morphological processing, the watershed algorithm is used to refine the segmentation of the vein pattern to avoid incorrect connections between veins; when performing binarization processing and connectivity analysis, the image after morphological processing is binarized by a preset threshold, the veins and the background are clearly distinguished, a binary image is obtained, the network structure belonging to the veins in the binary image is determined, and isolated pixels or small areas are removed to obtain a clean vein pattern; when performing hog feature extraction, the directional gradient histogram of the veins in the palm vein image is calculated to obtain the hog feature, and the shape, size and direction information of the veins are captured according to the hog feature.
[0041] Furthermore, the verification and identification module of the palm vein recognition login device based on a domestic CPU and operating system uses a nearest neighbor algorithm to perform feature matching verification, classifies the preprocessed palm vein image by comparing the similarity with the image in the palm vein image database, and compares it with a preset threshold to determine whether the palm vein image is successfully matched.
[0042] The benefits of the present invention are:
[0043] High security: Palm veins are located deep in the skin and are not easily stolen or forged. At the same time, the shape and size of the palm vary from person to person, making palm veins difficult to copy and imitate.
[0044] High accuracy: The shape and structure of palm veins are relatively stable and are not easily affected by environmental factors such as temperature and humidity.
[0045] Contactless: The palm vein recognition system does not require direct contact with the person being recognized. The recognition process can be completed by simply placing the palm on the recognition device, which is very convenient and hygienic.
[0046] Wide range of applications: The palm vein recognition system can be applied to various occasions, such as access control, identity authentication, payment, etc. At the same time, due to its high security and accuracy, it is also suitable for some fields with high security requirements.
[0047] Good repeatability: The morphology and structure of palmar veins are relatively stable and are not easily affected by factors such as age and disease. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 It is a schematic flow chart of the method of the present invention.
[0049] Figure 2 The application of the device of the present invention DETAILED DESCRIPTION
[0050] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.
[0051] Example 1
[0052] The present invention provides a palm vein recognition login method based on a domestically produced CPU and an operating system, comprising:
[0053] Step 1: Based on domestic CPU and operating system, input palm vein data:
[0054] Step 11: Use an infrared camera to capture the reflection image of the palm vein under infrared light to obtain a palm vein image.
[0055] The camera position and angle are preset to ensure that the palm vein pattern can be clearly captured. In addition, in order to adapt to the thickness and distance of different users' palms, the camera can also be equipped with autofocus function.
[0056] Step 2: Preprocess the palm vein image:
[0057] In step 2, when the palm vein image is denoised, Gaussian blur, median filtering, and bilateral filtering algorithms are used to remove noise.
[0058] When contrast enhancement is performed on palm vein images, the limited contrast adaptive histogram equalization method is used to increase the distinction between image regions and highlight the vein structure.
[0059] When resizing palm vein images, resize the palm vein images to a uniform size and resolution so that all images meet the same size standards.
[0060] When segmenting the palm vein image and performing rotation correction on the palm vein image, the palm vein area is separated from the background using the image segmentation method, and the palm vein image is angle-corrected to ensure consistency between images and facilitate subsequent feature matching.
[0061] The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent, that is, the image is simplified into black and white to make the vein pattern more prominent.
[0062] The segmented palm vein image can be further processed by morphology, such as a series of dilation and erosion operations, to remove small noise points and strengthen the vein pattern. Opening and closing operations are also often used to smooth the vein contour and separate the vein from the surrounding tissue. The segmented palm vein image is binarized and connected. The hog feature is extracted from the segmented palm vein image to construct a feature vector.
[0063] When further morphological processing is performed, the watershed algorithm can also be used to refine the segmentation of the vein pattern to avoid incorrect connections between veins; when performing binarization processing and connectivity analysis, the morphologically processed image is binarized by a preset threshold to clearly distinguish the veins from the background to obtain a binary image, determine the network structure belonging to the veins in the binary image, and remove isolated pixels or small areas to obtain a clean vein pattern; when performing hog feature extraction, the directional gradient histogram of the veins in the palm vein image is calculated to obtain the hog feature, and the shape, size and direction information of the veins are captured based on the hog feature.
[0064] Step 3: Verify palm vein data: Perform feature matching verification on the pre-processed palm vein image and the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
[0065] The nearest neighbor (nn) algorithm can be used for feature matching verification, and the palm vein image after preprocessing is classified by comparing the similarity with the image in the palm vein image database, and compared with the preset threshold to determine whether the palm vein image is matched successfully. The k-nearest neighbor (k-nn) is an extension of the nearest neighbor (nn) algorithm, which takes into account the voting of k nearest neighbors and score-based matching. The distance score between features, such as Euclidean distance, Hamming distance, etc., can be calculated, and compared with the preset threshold to determine whether the palm vein is matched successfully.
[0066] Example 2
[0067] The present invention provides a palm vein recognition login device based on a domestically produced CPU and an operating system, comprising an input module, a collection module, a preprocessing module and a verification and recognition module.
[0068] The input module is based on domestic CPU and operating system, and inputs palm vein data:
[0069] The collection module uses an infrared camera to capture the reflection image of the palm vein under infrared light to obtain the palm vein image.
[0070] The preprocessing module preprocesses the palm vein image:
[0071] De-noising of palm vein images.
[0072] Perform contrast enhancement on palm vein images.
[0073] Adjust the size of the palm vein image,
[0074] Segment the palm vein image and perform rotation correction on the palm vein image.
[0075] The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent.
[0076] Extracting features of relevant vein patterns of palm vein images;
[0077] The verification and recognition module verifies the palm vein data: the pre-processed palm vein image is feature matched with the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
[0078] As the information interaction and execution process between the modules of the above-mentioned device are based on the same concept as the embodiment of the method of the present invention, the specific contents can be found in the description of the embodiment of the method of the present invention and will not be repeated here.
[0079] Likewise, the benefits of the device of the present invention are:
[0080] High security: Palm veins are located deep in the skin and are not easily stolen or forged. At the same time, the shape and size of the palm vary from person to person, making palm veins difficult to copy and imitate.
[0081] High accuracy: The shape and structure of palm veins are relatively stable and are not easily affected by environmental factors such as temperature and humidity.
[0082] Contactless: The palm vein recognition system does not require direct contact with the person being recognized. The recognition process can be completed by simply placing the palm on the recognition device, which is very convenient and hygienic.
[0083] Wide range of applications: The palm vein recognition system can be applied to various occasions, such as access control, identity authentication, payment, etc. At the same time, due to its high security and accuracy, it is also suitable for some fields with high security requirements.
[0084] Good repeatability: The morphology and structure of palmar veins are relatively stable and are not easily affected by factors such as age and disease.
[0085] It should be noted that not all steps and modules in the above-mentioned processes and device structures are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.
[0086] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.
Claims
1. A palm vein recognition login method based on a domestic CPU and operating system, characterized in that include: Step 1: Based on domestic CPU and operating system, input palm vein data: Step 11: Use an infrared camera to capture the reflection image of the palm vein under infrared light to obtain a palm vein image. Step 2: Preprocess the palm vein image: De-noising of palm vein images. Perform contrast enhancement on palm vein images. Adjust the size of the palm vein image, Segment the palm vein image and perform rotation correction on the palm vein image. The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent. Extracting features of relevant vein patterns of palm vein images; Step 3: Verify palm vein data: Perform feature matching verification on the pre-processed palm vein image and the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
2. According to the palm vein recognition login method based on a domestic CPU and an operating system according to claim 1, it is characterized in that when the palm vein image is denoised in step 2, Gaussian blur, median filtering, and bilateral filtering algorithms are used to remove noise. When contrast enhancement is performed on palm vein images, the limited contrast adaptive histogram equalization method is used to increase the distinction between image regions and highlight the vein structure. When resizing palm vein images, resize the palm vein images to a uniform size and resolution so that all images meet the same size standards. When segmenting the palm vein image and performing rotation correction on the palm vein image, the palm vein area is separated from the background using the image segmentation method, and the palm vein image is angle-corrected to ensure consistency between images and facilitate subsequent feature matching.
3. According to claim 2, a palm vein recognition login method based on a domestic CPU and an operating system is characterized in that When the palm vein image is segmented in step 2, it also includes: performing morphological processing on the segmented palm vein image, performing binarization processing and connectivity analysis on the segmented palm vein image, performing hog feature extraction on the segmented palm vein image, and constructing a feature vector.
4. According to claim 3, a palm vein recognition login method based on a domestic CPU and an operating system, The method is characterized in that when performing morphological processing in step 2, a watershed algorithm is used to refine the segmentation of the vein pattern to avoid incorrect connections between veins; when performing binarization processing and connectivity analysis, the image after morphological processing is binarized by a preset threshold value to clearly distinguish the veins from the background to obtain a binary image, determine the network structure belonging to the veins in the binary image, and remove isolated pixels or small areas to obtain a clean vein pattern; When extracting the hog feature, the directional gradient histogram of the veins in the palm vein image is calculated to obtain the hog feature, and the shape, size and direction information of the veins are captured according to the hog feature.
5. According to claim 1, a palm vein recognition login method based on a domestic CPU and an operating system is characterized in that In step 3, the nearest neighbor algorithm is used to perform feature matching verification. The palm vein image after preprocessing is classified by comparing the similarity with the images in the palm vein image database, and is compared with a preset threshold to determine whether the palm vein image is successfully matched.
6. A palm vein recognition login device based on a domestically produced CPU and operating system, characterized in that It includes input module, collection module, pre-processing module and verification and recognition module. The input module is based on domestic CPU and operating system, and inputs palm vein data: The collection module uses an infrared camera to capture the reflection image of the palm vein under infrared light to obtain the palm vein image. The preprocessing module preprocesses the palm vein image: De-noising of palm vein images. Perform contrast enhancement on palm vein images. Adjust the size of the palm vein image, Segment the palm vein image and perform rotation correction on the palm vein image. The palm vein image is grayscale normalized and binarized to make the vein pattern more prominent. Extracting features of relevant vein patterns of palm vein images; The verification and recognition module verifies the palm vein data: the pre-processed palm vein image is feature matched with the image in the palm vein image database. If the match is successful, the user is authorized to access the system, otherwise the access is denied and the user is prompted to re-authenticate.
7. According to claim 6, a palm vein recognition and login device based on a domestically produced CPU and an operating system is characterized in that When the preprocessing module performs denoising on the palm vein image, it uses Gaussian blur, median filtering, and bilateral filtering algorithms to remove noise. When contrast enhancement is performed on palm vein images, the limited contrast adaptive histogram equalization method is used to increase the distinction between image regions and highlight the vein structure. When resizing palm vein images, resize the palm vein images to a uniform size and resolution so that all images meet the same size standards. When segmenting the palm vein image and performing rotation correction on the palm vein image, the palm vein area is separated from the background using the image segmentation method, and the palm vein image is angle-corrected to ensure consistency between images and facilitate subsequent feature matching.
8. According to claim 7, a palm vein recognition and login device based on a domestically produced CPU and an operating system is characterized in that When the preprocessing module segments the palm vein image, it also includes: performing morphological processing on the segmented palm vein image, performing binarization processing and connectivity analysis on the segmented palm vein image, performing hog feature extraction on the segmented palm vein image, and constructing a feature vector.
9. A palm vein recognition login device based on a domestically produced CPU and operating system according to claim 8, characterized in that When the preprocessing module performs morphological processing, the watershed algorithm is used to refine the segmentation of the vein pattern to avoid incorrect connections between veins; when performing binarization processing and connectivity analysis, the image after morphological processing is binarized by a preset threshold to clearly distinguish the veins from the background to obtain a binary image, determine the network structure belonging to the veins in the binary image, and remove isolated pixels or small areas to obtain a clean vein pattern; When extracting the hog feature, the directional gradient histogram of the veins in the palm vein image is calculated to obtain the hog feature, and the shape, size and direction information of the veins are captured according to the hog feature.
10. The palm vein recognition login device based on a domestically produced CPU and operating system according to claim 6, characterized in that The verification and recognition module uses the nearest neighbor algorithm to perform feature matching verification, classifies the preprocessed palm vein image by comparing its similarity with the images in the palm vein image database, and compares it with the preset threshold to determine whether the palm vein image is successfully matched.