Infrared palm living body detection method, system and device based on high frequency and storage medium

Abstract

The invention discloses an infrared palm living body detection method, system and device based on high frequency and a storage medium. The method comprises the steps that S1, an infrared palm image is acquired; s2, Fourier transform is carried out on the infrared palm image to obtain frequency spectrum information, and a high-frequency part is extracted; s3, filtering the high-frequency part to enhance detail information such as blood vessels in the image; s4, performing Fourier inverse transformation on the high-frequency part to obtain a high-frequency infrared palm image; and S5, carrying out living body recognition on the high-frequency infrared palm image by using a living body recognition model. According to the invention, the palm can be accurately subjected to living body recognition.

Description

Background Art

[0002] With the rapid development of biometric technologies, as a key link to ensure the security of biometric recognition systems, live detection technology has been increasingly emphasized. Among numerous biometric features, palm features have become one of the research hotspots in the biometric field due to their uniqueness and stability. However, traditional palm recognition technologies mainly rely on visible light images and are vulnerable to factors such as lighting conditions and spoofing attacks (such as using fake hands or photos), resulting in a decline in recognition accuracy and security.

[0003] To overcome these challenges, researchers have started to explore palm live detection technologies based on infrared imaging. Infrared imaging technology captures images by utilizing the differences in the surface temperature distribution of objects and has significant advantages for live detection. On the one hand, due to blood circulation and metabolic activities, the surface temperature distribution of a live palm is significantly different from that of a fake hand or a photo; on the other hand, infrared imaging is not restricted by lighting conditions and can operate stably in various lighting environments.

[0004] However, directly using infrared palm images for live detection still faces some difficulties. High-frequency components in infrared images, such as vascular networks and fine textures, are key features for distinguishing live palms from forged ones. However, these high-frequency components are often relatively weak and vulnerable to noise interference, making it difficult to directly use them for live recognition. Therefore, how to effectively extract and utilize the high-frequency components in infrared palm images has become the key to improving the accuracy and robustness of live detection.

[0005] The disclosure of the above background art content is only used to assist in understanding the inventive concept and technical solution of the present invention, and it does not necessarily belong to the prior art of this patent application. Without clear evidence indicating that the above content was publicly available on the filing date of this patent application, the above background art should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0006] For this reason, the present invention proposes a high-frequency-based infrared palm live detection method. By performing Fourier transform to extract the high-frequency part of the infrared palm image, using filtering processing to enhance detailed information such as blood vessels in the image, then performing inverse Fourier transform to reconstruct the high-frequency infrared palm image, and using a live recognition model for recognition and judgment, it can not only effectively extract and utilize the high-frequency components in the infrared palm image, improve the accuracy and robustness of live detection, but also adapt to various lighting environments and spoofing attack scenarios, providing a strong guarantee for the security of biometric recognition systems.

[0007] In a first aspect, the present invention provides a high-frequency-based infrared palm live detection method, characterized by including:

[0008] Step S1: Obtain an infrared palm image;

[0009] Step S2: Perform Fourier transform on the infrared palm image to obtain spectral information and extract the high-frequency part;

[0010] Step S3: Perform filtering on the high-frequency part to enhance detailed information such as blood vessels in the image;

[0011] Step S4: Perform inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image;

[0012] Step S5: Use a live detection model to perform live detection on the high-frequency infrared palm image.

[0013] Optionally, in the above-mentioned infrared palm live detection method based on high frequency, Step S2 includes:

[0014] Step S21: Perform two-dimensional Fourier transform on the infrared palm image;

[0015] Step S22: Analyze the result of the two-dimensional Fourier transform to obtain spectral information;

[0016] Step S23: Extract the high-frequency part according to the spectral information.

[0017] Optionally, in the above-mentioned infrared palm live detection method based on high frequency, Step S3 includes:

[0018] Step S31: Select a high-frequency filter according to the characteristics of high-frequency components;

[0019] Step S32: Apply the high-frequency filter to the high-frequency part.

[0020] Optionally, in the above-mentioned infrared palm live detection method based on high frequency, Step S4 includes:

[0021] Step S41: Perform inverse two-dimensional Fourier transform on the high-frequency part;

[0022] Step S42: Reconstruct a high-frequency infrared palm image according to the result after the inverse two-dimensional Fourier transform.

[0023] Optionally, in the above-mentioned infrared palm live detection method based on high frequency, Step S5 includes:

[0024] Step S51: Extract features from the high-frequency infrared palm image to obtain high-frequency palm features;

[0025] Step S52: Input the high-frequency palm features into the live detection model to obtain a live detection result.

[0026] Optionally, in the infrared palm liveness detection method based on high frequency, during the training of the liveness recognition model, the infrared palm images are used for supervision.

[0027] Optionally, in the infrared palm liveness detection method based on high frequency, during the training of the liveness recognition model, by setting different parameters, different high-frequency parts are obtained to improve the adaptability of the liveness recognition model.

[0028] In a second aspect, the present invention provides an infrared palm liveness detection system based on high frequency, which is used to implement the infrared palm liveness detection method based on high frequency described in any one of the foregoing items. The system includes:

[0029] An acquisition module, configured to acquire infrared palm images;

[0030] A transformation module, configured to perform Fourier transform on the infrared palm images to obtain spectral information and extract the high-frequency part;

[0031] A filtering module, configured to perform filtering processing on the high-frequency part to enhance detailed information such as blood vessels in the image;

[0032] An inverse transformation module, configured to perform inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image;

[0033] A recognition module, configured to use a liveness recognition model to perform liveness recognition on the high-frequency infrared palm image.

[0034] In a third aspect, the present invention provides an infrared palm liveness detection device based on high frequency, which includes:

[0035] A processor;

[0036] A memory, which stores executable instructions of the processor;

[0037] Wherein, the processor is configured to execute the steps of the infrared palm liveness detection method based on high frequency described in any one of the foregoing items by executing the executable instructions.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium for storing a program, and the program, when executed, implements the steps of the infrared palm liveness detection method based on high frequency described in any one of the foregoing items.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention extracts the high-frequency part of the infrared palm image through Fourier transform. This step focuses on the fine structures and texture features in the image, especially the physiological features unique to the living body, such as the blood vessel network. These high-frequency features play a crucial role in live detection because they can reflect the true physiological state of the organism and effectively distinguish between the living body and forged samples.

[0041] The present invention performs filtering on the high-frequency part, which not only removes noise interference but also enhances the visibility of key detail information such as blood vessels. This step significantly improves the image quality, enabling the live recognition model to more accurately capture the key features for distinguishing the living body from the forgery, thereby improving the recognition accuracy and robustness.

[0042] The infrared image of the living palm in the present invention contains rich physiological information, especially the temperature changes generated by blood circulation. These information are difficult to replicate for forgeries (such as photos, silicone hand models, etc.). Therefore, this method can effectively identify and resist various camouflage attacks and improve the security of the biometric system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings. By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, objectives, and advantages of the present invention will become more obvious:

[0044] Figure 1 It is a flowchart of the steps of a high-frequency-based infrared palm live detection method in an embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of a high-frequency image in an embodiment of the present invention;

[0046] Figure 3 It is a schematic diagram of a frequency spectrum distribution in an embodiment of the present invention;

[0047] Figure 4 It is a flowchart of the steps of obtaining the high-frequency part in an embodiment of the present invention;

[0048] Figure 5 It is a flowchart of the steps of performing filtering in an embodiment of the present invention;

[0049] Figure 6 It is a flowchart of the steps of obtaining a high-frequency infrared palm image in an embodiment of the present invention;

[0050] Figure 7 It is a flowchart of steps for performing live body recognition in an embodiment of the present invention;

[0051] Figure 8 It is a schematic structural diagram of an infrared palm live body detection system based on high frequency in an embodiment of the present invention;

[0052] Figure 9 It is a schematic structural diagram of an infrared palm live body detection device based on high frequency in an embodiment of the present invention; and

[0053] Figure 10 It is a schematic structural diagram of a computer-readable storage medium in an embodiment of the present invention. Specific embodiments

[0054] The present invention will be described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those of ordinary skill in the art can make several deformations and improvements without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0055] The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here, for example, can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any of their deformations are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0056] An infrared palm live body detection method based on high frequency provided by an embodiment of the present invention aims to solve the problems existing in the prior art.

[0057] The technical solutions of the present invention and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the drawings.

[0058] The present invention proposes a high-frequency-based infrared palm liveness detection method. By performing Fourier transform to extract the high-frequency part in the infrared palm image, and using filtering processing to enhance the detailed information such as blood vessels in the image, then reconstructing the high-frequency infrared palm image through inverse Fourier transform, and using a liveness recognition model for recognition and judgment. It can not only effectively extract and utilize the high-frequency components in the infrared palm image, improve the accuracy and robustness of liveness detection, but also adapt to various lighting environments and camouflage attack scenarios, providing a strong guarantee for the security of biometric recognition systems.

[0059] Figure 1 It is a flowchart of the steps of a high-frequency-based infrared palm liveness detection method in an embodiment of the present invention. As Figure 1 shown, the steps of a high-frequency-based infrared palm liveness detection method in an embodiment of the present invention include:

[0060] Step S1: Obtain an infrared palm image.

[0061] In this step, an infrared image of the palm is captured by a dedicated infrared camera. Since the blood vessel network inside the palm has specific absorption and reflection characteristics for near-infrared light, the infrared camera can capture these subtle physiological characteristics. The obtained infrared palm image should contain sufficient information for extraction and analysis in subsequent steps.

[0062] Step S2: Perform Fourier transform on the infrared palm image to obtain spectral information and extract the high-frequency part.

[0063] In this step, Fourier transform is a mathematical method for converting an image in the spatial domain into an image in the frequency domain. In this step, Fourier transform is performed on the infrared palm image to obtain its spectral information. The spectral information reflects the distribution of different frequency components in the image. The high-frequency part usually corresponds to the detailed information such as edges and textures in the image, while the low-frequency part corresponds to the overall structure and background information of the image.

[0064] By extracting the high-frequency part, it is possible to focus on the subtle structures and texture features in the image, especially the physiological features unique to the living body such as the blood vessel network. These high-frequency features play a crucial role in liveness detection because they can reflect the true physiological state of the living body, thus effectively distinguishing between the living body and forged samples.

[0065] Figure 2 Images of different spectra of the palm are shown. In Figure 2 , "Orig" represents the original image, "lowfrerquency" represents the low-spectrum image, and "high frequency" represents the high-frequency image. It can be seen from the figure that the high-frequency part obviously has characteristics different from the low-spectrum part and the original image.

[0066] Figure 3 is the spectral distribution. Figure 3 In it, "Low Frequency" represents the spectral distribution of the low-frequency part, and "High Frequency" represents the spectral distribution of the high-frequency part. From Figure 3 it can be seen that the distributions of the two are significantly different and can be used for live body recognition.

[0067] Step S3: Perform filtering processing on the high-frequency part to enhance detailed information such as blood vessels in the image.

[0068] In this step, after extracting the high-frequency part, in order to further enhance detailed information such as blood vessels in the image, it is necessary to perform filtering processing on these high-frequency components. The purpose of the filtering processing is to remove noise interference while retaining and enhancing key detailed information. In this step, various filtering methods can be used, such as Gaussian filtering, median filtering, or adaptive filtering, etc. Select appropriate filtering methods and parameters according to specific situations to achieve the best filtering effect. Through the filtering processing, detailed information such as blood vessels can be made more clearly distinguishable in the image, thereby improving the accuracy of live body recognition.

[0069] Step S4: Perform inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image.

[0070] In this step, after completing the filtering processing, it is necessary to perform inverse Fourier transform on the high-frequency part to convert it from the frequency domain back to the spatial domain. This step is the inverse process of the Fourier transform, and through the inverse Fourier transform, the processed high-frequency infrared palm image can be obtained. The high-frequency infrared palm image mainly contains enhanced detailed information such as blood vessels. These information play a key role in live body recognition because they can reflect the true physiological characteristics of the organism, thereby helping the recognition system accurately determine whether the palm is a live body.

[0071] Step S5: Use the live body recognition model to perform live body recognition on the high-frequency infrared palm image.

[0072] In this step, use the live body recognition model to perform live body recognition on the high-frequency infrared palm image. The live body recognition model is usually constructed based on deep learning or machine learning algorithms, which can automatically extract and analyze the feature information in the image and determine whether the palm is a live body. In this step, input the high-frequency infrared palm image into the live body recognition model, and the model will perform processing such as feature extraction, classification, and decision-making on the image. Finally, the model will output a recognition result to determine whether the palm is a live body. According to the recognition result, corresponding security responses or control measures can be made.

[0073] Figure 4 is the flowchart of the steps for obtaining the high-frequency part in the embodiment of the present invention. AsFigure 4 As shown in Figure 4 , the steps of obtaining the high-frequency part in the embodiments of the present invention include:

[0074] Step S21: Perform two-dimensional Fourier transform on the infrared palm image.

[0075] In this step, the infrared palm image is transformed from the spatial domain to the frequency domain by using two-dimensional Fourier transform (2D Fourier Transform). Two-dimensional Fourier transform is a mathematical tool that can decompose an image into a combination of sine waves and cosine waves with different frequencies and directions. The amplitude and phase information of these waves constitute the spectrum of the image.

[0076] Specifically, the infrared palm image is regarded as a two-dimensional matrix, where each element represents the pixel value at a certain position in the image. Then, the two-dimensional Fourier transform algorithm is applied to process this matrix to obtain a new matrix, that is, the spectrum matrix of the image. Each element in this matrix corresponds to the amplitude and phase of a certain frequency component in the image.

[0077] Step S22: Analyze the result of the two-dimensional Fourier transform to obtain spectrum information.

[0078] In this step, after obtaining the spectrum matrix, it is necessary to analyze it to obtain spectrum information. Spectrum information reflects the distribution of different frequency components in the image. In the result of two-dimensional Fourier transform, the low-frequency components are usually located near the center of the spectrum matrix, while the high-frequency components are distributed at the edges of the matrix.

[0079] By analyzing the spectrum matrix, information such as the relative intensity, distribution range of the low-frequency and high-frequency components in the image, and the proportional relationship between them can be obtained. These information are very important for subsequent extraction of the high-frequency part and live body recognition.

[0080] Step S23: Extract the high-frequency part according to the spectrum information.

[0081] In this step, after obtaining the spectrum information, the high-frequency part in the image can be extracted according to this information. The high-frequency part usually corresponds to the details such as edges and textures in the image, and these information play a crucial role in live body detection.

[0082] There are various methods for extracting the high-frequency part. One common method is to set a frequency threshold. According to the distribution of the spectrum information, a suitable frequency threshold can be set. Then, the frequency components in the spectrum matrix higher than this threshold are retained, while the frequency components lower than this threshold are set to zero or other processing is performed. In this way, a spectrum matrix containing only high-frequency components can be obtained.

[0083] Finally, by performing a two-dimensional inverse Fourier transform (2D Inverse Fourier Transform) on this high-frequency spectrum matrix, it can be converted back to the spatial domain to obtain an infrared palm image that only contains high-frequency information. This high-frequency image will be used as the input for liveness detection in subsequent steps.

[0084] In this embodiment, through steps such as two-dimensional Fourier transform, spectrum information analysis, and high-frequency part extraction, the high-frequency information in the infrared palm image is successfully extracted, providing important feature information for subsequent liveness detection.

[0085] Figure 5 It is a flowchart of the steps for performing filtering processing in an embodiment of the present invention. As Figure 5 shown, the steps for performing filtering processing in an embodiment of the present invention include:

[0086] Step S31: Select a high-frequency filter according to the characteristics of high-frequency components.

[0087] In this step, it is necessary to select a suitable filter according to the characteristics of high-frequency components. High-frequency components usually correspond to detailed information such as edges and textures in the image, and these information play a crucial role in liveness detection. Therefore, a filter that can retain and enhance these high-frequency detailed information needs to be selected.

[0088] There are various types of high-frequency filters, such as high-pass filters, band-pass filters, and edge enhancement filters, etc. A high-pass filter allows components above a certain frequency to pass through, while components below that frequency are suppressed. A band-pass filter only allows components within a specific frequency range to pass through. An edge enhancement filter is specifically used to enhance the edge information in the image.

[0089] When selecting a filter, it is necessary to consider the frequency range, distribution characteristics of high-frequency components, and the characteristics of the filter. For example, if the high-frequency components are mainly concentrated in a specific frequency range, a band-pass filter can be selected to retain the components within this range. If it is necessary to enhance the edge information in the image, an edge enhancement filter can be selected.

[0090] Step S32: Apply the high-frequency filter to the high-frequency part.

[0091] In this step, it needs to be applied to the extracted high-frequency part. This step is usually implemented through a convolution operation. The convolution operation is a mathematical operation that obtains the filtered image by multiplying the filter (also called the convolution kernel) with the image point by point and summing.

[0092] During specific operations, the high-frequency filter is used as the convolution kernel, and the extracted high-frequency part is used as the input image for convolution operations. The result of the convolution operation is a filtered high-frequency image. The high-frequency components in this image are retained and enhanced, while noise and other unwanted components are suppressed or removed. It is possible to further highlight the high-frequency detail information in the image, such as the vascular network, etc., thereby providing more accurate and reliable feature information for subsequent live body recognition.

[0093] In this embodiment, by selecting a suitable high-frequency filter and applying it to the high-frequency part, the filtering process of the high-frequency components is successfully achieved, providing clearer and more accurate feature information for subsequent live body recognition.

[0094] Figure 6 It is a flowchart of the steps for obtaining a high-frequency infrared palm image in an embodiment of the present invention. As Figure 6 shown, the steps for obtaining a high-frequency infrared palm image in an embodiment of the present invention include:

[0095] Step S41: Perform two-dimensional inverse Fourier transform on the high-frequency part.

[0096] In this step, perform two-dimensional inverse Fourier transform (2D Inverse Fourier Transform) on the filtered high-frequency part. The two-dimensional inverse Fourier transform is the inverse process of the two-dimensional Fourier transform, which can convert the spectral information of the image back to the spatial domain information, that is, reconstruct the original image or the image corresponding to its specific frequency components.

[0097] During specific operations, take the filtered high-frequency part (represented as a spectral matrix in the frequency domain) as the input and apply the two-dimensional inverse Fourier transform algorithm for calculation. This algorithm will perform inverse transform processing on each element in the spectral matrix, and finally obtain a two-dimensional matrix. Each element in this matrix corresponds to the pixel value at a certain position in the spatial domain. This two-dimensional matrix is the representation of the required high-frequency infrared palm image in the spatial domain.

[0098] It should be noted that when performing the two-dimensional inverse Fourier transform, it may encounter the situation of spectral matrix centering (i.e., moving the low-frequency components to the center of the spectrum) or non-centering. If the spectral matrix is centered, the inverse-transformed image will directly correspond to the high-frequency part of the original image; if the spectral matrix is not centered, it may be necessary to perform centering processing before the inverse transform to ensure that the inverse-transformed image correctly corresponds to the high-frequency part of the original image.

[0099] Step S42: Reconstruct the high-frequency infrared palm image according to the result of the two-dimensional inverse Fourier transform.

[0100] In this step, after obtaining the result of the inverse two-dimensional Fourier transform, it is necessary to reconstruct the high-frequency infrared palm image based on this result. This image mainly contains the high-frequency components in the original infrared palm image, such as details like edges and textures.

[0101] During the specific operation, the two-dimensional matrix after the inverse two-dimensional Fourier transform (i.e., the spatial domain representation of the high-frequency part) can be converted into an image format. This image format can be common image file types, such as BMP, JPEG, PNG, etc. During the conversion process, it is necessary to set information such as the color and brightness of each pixel in the image according to the pixel values in the matrix.

[0102] The reconstructed high-frequency infrared palm image will be used as the input for subsequent liveness recognition. Since this image mainly contains high-frequency components, it can more clearly display details such as the blood vessel network in the palm, thereby improving the accuracy and reliability of liveness recognition.

[0103] In this embodiment, through steps such as the inverse two-dimensional Fourier transform and image reconstruction, the high-frequency part is successfully converted from the frequency domain back to the spatial domain, and the high-frequency infrared palm image is reconstructed. This image provides important feature information for subsequent liveness recognition.

[0104] Figure 7 It is a flowchart of the steps for liveness recognition in an embodiment of the present invention. As Figure 7 shown, the steps for liveness recognition in an embodiment of the present invention include:

[0105] Step S51: Extract features from the high-frequency infrared palm image to obtain high-frequency palm features.

[0106] In this step, it is necessary to extract distinguishable features from the high-frequency infrared palm image, and these features will be used for subsequent liveness recognition. The high-frequency infrared palm image mainly contains details such as the blood vessel network and texture of the palm, and these information play a crucial role in liveness recognition.

[0107] There are various methods for feature extraction, and it can be selected according to specific application scenarios and requirements. The following are some common feature extraction methods:

[0108] Edge detection: Extract features by detecting the edge information in the image. Edges are the places where the gray value changes most violently in the image, usually corresponding to the contours of objects. Edge detection algorithms such as Canny edge detection and Sobel operator can be used to extract edge features.

[0109] Texture analysis: Extract features by analyzing the texture information in the image. Texture is the local pattern that appears repeatedly in the image, and texture features can be extracted through algorithms such as gray-level co-occurrence matrix (GLCM) and local binary pattern (LBP).

[0110] Shape analysis: Extract features by analyzing the shape of the object in the image. Shape descriptors (such as contour, area, perimeter, etc.) can be used to describe the shape features of the object.

[0111] In high-frequency infrared palm images, the vascular network is an important feature. The vascular network features can be extracted through image processing techniques (such as threshold segmentation, morphological processing, etc.). In addition, other feature extraction methods, such as texture analysis and shape analysis, can be combined to extract more feature information.

[0112] The extracted high-frequency palm features will be used as the input for the subsequent live body recognition model. These features should be robust, stable, and discriminative to ensure the accuracy and reliability of live body recognition.

[0113] Step S52: Input the high-frequency palm features into the live body recognition model to obtain the live body recognition result.

[0114] In this step, the extracted high-frequency palm features are input into the live body recognition model to obtain the live body recognition result. The live body recognition model is a trained machine learning model that can judge whether the palm is a live body based on the input feature information.

[0115] The live body recognition model can be implemented using various algorithms, such as support vector machine (SVM), neural network (NN), convolutional neural network (CNN), etc. These algorithms can all extract useful information from the input feature information and perform classification judgments.

[0116] Before inputting the high-frequency palm features into the live body recognition model, some preprocessing operations may be required, such as feature normalization, dimensionality reduction, etc. These operations can improve the performance and stability of the model.

[0117] The result output by the live body recognition model is usually a probability value or a classification label, indicating the possibility that the input palm features are a live body or the classification result. According to the specific application scenario and requirements, different thresholds can be set to judge the result of live body recognition. If the probability value is higher than the threshold, it is judged as a live body; otherwise, it is judged as a non-live body.

[0118] This embodiment successfully realizes the processing and recognition of high-frequency infrared palm images through steps such as feature extraction and live body recognition model. This process provides important feature information and judgment basis for live body recognition.

[0119] In some embodiments, when the living body recognition model is trained, the infrared palm image is used for supervision. Supervised learning is a method in machine learning that uses labeled training data to train a model. During the training process, the model learns how to map input data (features) to output data (labels). When the model receives new input data, it can use the learned mapping relationship to predict the output.

[0120] The role of the infrared palm image in supervised learning is reflected in feature extraction, label annotation, and model training.

[0121] For feature extraction, the infrared palm image provides rich feature information, such as blood vessel networks, textures, etc. These information are crucial for living body recognition. During the training process, the model learns how to extract these features from the infrared palm image.

[0122] For label annotation, each infrared palm image requires a corresponding label to indicate whether the image is a living body. These labels are usually obtained through manual annotation or existing databases. During the training process, the model uses these labels to learn how to distinguish between living bodies and non-living bodies.

[0123] For model training, the extracted features and the corresponding labels are input into the model, and the parameters of the model are continuously adjusted through an iterative optimization algorithm (such as gradient descent) so that the prediction results of the model for the input data are closer and closer to the true labels.

[0124] Using the infrared palm image for supervision in this embodiment is an effective method for training a living body recognition model. By reasonable data preprocessing, model selection, and parameter tuning, a living body recognition model with high performance and generalization ability can be trained.

[0125] In some embodiments, when the living body recognition model is trained, by setting different parameters, different high-frequency parts are obtained to improve the adaptability of the living body recognition model. The core of this embodiment is to use image processing technology to extract the high-frequency information in the image and use this information as features to input into the model to enhance the model's ability to distinguish between living bodies and non-living bodies.

[0126] High-frequency information usually refers to the parts in the image that change rapidly, and these parts often contain rich detail and texture features. In living body recognition, high-frequency information may include fine features such as the blood vessel network and skin texture of the palm, and there are usually significant differences in these features between living bodies and non-living bodies. To extract high-frequency information, various image processing technologies can be used, such as edge detection algorithms like Laplace transform and Sobel operator, or frequency domain analysis (such as Fourier transform) to separate the high-frequency components of the image. These algorithms can capture the subtle changes in the image and thus extract high-frequency features.

[0127] When training a live body recognition model, the extraction process of high-frequency information can be adjusted by setting different parameters. These parameters may include the size, shape, direction of the filter, etc., as well as the specific implementation details of the image processing algorithm. By adjusting these parameters, high-frequency information in different frequency ranges can be obtained, thereby enriching the input feature set of the model.

[0128] In the model training stage, a labeled training data set is required to guide the learning process of the model. The training data set should contain a sufficient number of live and non-live samples to ensure that the model can learn effective feature representations. At the same time, the model also needs to be fully iteratively trained to gradually converge and achieve a high recognition accuracy.

[0129] By setting different parameters and extracting different high-frequency information, the live body recognition model can have stronger adaptability to different types of input images. Specifically, the model can learn more diverse feature representations, so that it can still maintain good recognition performance when facing complex scenarios such as different lighting conditions, different angles, and different resolutions.

[0130] In addition, this method of parameter adjustment also helps to improve the robustness and generalization ability of the model. In practical applications, the live body recognition model may encounter various unknown or unpredictable input situations. By setting different parameters during training and extracting diverse high-frequency information, the model can better handle these unknown situations, thereby improving the stability and reliability of recognition.

[0131] In this embodiment, by setting different parameters and extracting different high-frequency information, the adaptability of the live body recognition model can be significantly improved. This method helps the model learn more diverse feature representations, so that it can still maintain good recognition performance when facing complex scenarios.

[0132] Figure 8 It is a schematic structural diagram of an infrared palm live body detection system based on high frequency in an embodiment of the present invention. As Figure 8 shown, an infrared palm live body detection system based on high frequency in an embodiment of the present invention includes:

[0133] An acquisition module, configured to acquire an infrared palm image;

[0134] A transformation module, configured to perform a Fourier transform on the infrared palm image to obtain spectral information and extract the high-frequency part;

[0135] A filtering module, configured to perform filtering processing on the high-frequency part to enhance detail information such as blood vessels in the image;

[0136] An inverse transformation module, configured to perform an inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image;

[0137] An identification module for performing liveness identification on the high-frequency infrared palm image by using a liveness identification model.

[0138] Specifically, the acquisition module is responsible for capturing or receiving an infrared palm image. This is the starting point of the entire detection process and provides the basic data for subsequent processing.

[0139] The transformation module performs a Fourier transform on the acquired infrared palm image. The Fourier transform is a mathematical tool that can transform an image from the spatial domain to the frequency domain to analyze the frequency components of the image. The purpose of this module is to extract the high-frequency part from the transformed spectrum information. These high-frequency parts usually contain detailed information in the image, such as the vascular network, etc.

[0140] The filtering module performs filtering on the high-frequency part extracted by the transformation module. The purpose of filtering is to enhance specific detailed information in the image, such as blood vessels, etc., and may suppress or remove unwanted frequency components. This helps to improve the accuracy of liveness identification. The filtering module depends on the high-frequency information provided by the transformation module. Through precise filtering, the liveness features can be further highlighted and more valuable information can be provided for the subsequent identification module.

[0141] The inverse transformation module performs an inverse Fourier transform on the filtered high-frequency part, converting it from the frequency domain back to the spatial domain to generate a high-frequency infrared palm image. This step is crucial for restoring the spatial information of the image, enabling the identification module to directly process and analyze the image. The inverse transformation module is closely connected to the transformation module and the filtering module. It receives the filtered high-frequency information and converts it back into an image format that can be used for identification.

[0142] The identification module uses a liveness identification model to identify the high-frequency infrared palm image generated by the inverse transformation module. The identification module may be constructed based on deep learning, machine learning, or other algorithms and is used to distinguish between a live palm and a non-live palm. The identification module is the final link in the entire system and depends on the preprocessed images provided by the previous modules. Through a well-trained model, the identification module can accurately determine whether the palm in the image is a live one.

[0143] This embodiment proposes a high-frequency-based infrared palm liveness detection method. By extracting the high-frequency part of the infrared palm image through Fourier transform, enhancing detailed information such as blood vessels in the image through filtering, reconstructing the high-frequency infrared palm image through inverse Fourier transform, and using a liveness identification model for identification and judgment, it can not only effectively extract and utilize the high-frequency components in the infrared palm image, improve the accuracy and robustness of liveness detection, but also adapt to various lighting environments and camouflage attack scenarios, providing a strong guarantee for the security of biometric recognition systems.

[0144] In an embodiment of the present invention, there is also provided an infrared palm liveness detection device based on high frequency, including a processor and a memory storing executable instructions of the processor. The processor is configured to execute the steps of an infrared palm liveness detection method based on high frequency via executing the executable instructions.

[0145] As described above, this embodiment proposes an infrared palm liveness detection method based on high frequency. By performing Fourier transform to extract the high-frequency part in the infrared palm image, and using filtering processing to enhance details such as blood vessels in the image, then reconstructing the high-frequency infrared palm image through inverse Fourier transform, and using a liveness recognition model for recognition and judgment. It can not only effectively extract and utilize the high-frequency components in the infrared palm image, improve the accuracy and robustness of liveness detection, but also adapt to various lighting environments and camouflage attack scenarios, providing a strong guarantee for the security of the biometric recognition system.

[0146] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, a method, or a program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "platform" here.

[0147] Figure 9 It is a schematic structural diagram of an infrared palm liveness detection device based on high frequency in an embodiment of the present invention. The following will refer to Figure 9 to describe the electronic device 600 according to this embodiment of the present invention. Figure 9 The electronic device 600 shown is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0148] As Figure 9 shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0149] Among them, the storage unit stores program code, and the program code can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present invention described in the above part of the infrared palm liveness detection method based on high frequency in this specification. For example, the processing unit 610 can execute the steps as Figure 1 shown.

[0150] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

[0151] The storage unit 620 may also include a program / utilities 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a grid environment.

[0152] The bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of a variety of bus structures.

[0153] The electronic device 600 may also communicate with one or more external devices 700 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or may communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through the input / output (I / O) interface 650. And, the electronic device 600 may also communicate with one or more grids (such as a local area network (LAN), a wide area network (WAN), and / or a public grid, such as the Internet) through the grid adapter 660. The grid adapter 660 may communicate with other modules of the electronic device 600 through the bus 630. It should be understood that although Figure 9 not shown, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0154] An embodiment of the present invention also provides a computer-readable storage medium for storing a program, and when the program is executed, it implements the steps of a high-frequency-based infrared palm liveness detection method. In some possible implementation manners, various aspects of the present invention may also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above part of the high-frequency-based infrared palm liveness detection method of this specification.

[0155] As shown above, this embodiment proposes an infrared palm liveness detection method based on high frequency. The high frequency part in the infrared palm image is extracted through Fourier transform, and the details such as blood vessels in the image are enhanced by filtering processing. Then, the high frequency infrared palm image is reconstructed through inverse Fourier transform, and the liveness recognition model is used for recognition and judgment. It can not only effectively extract and utilize the high frequency components in the infrared palm image, improve the accuracy and robustness of liveness detection, but also adapt to various lighting environments and camouflage attack scenarios, providing a strong guarantee for the security of the biometric recognition system.

[0156] Figure 10 It is a schematic structural diagram of the computer-readable storage medium in the embodiment of the present invention. Refer to Figure 10 As shown, a program product 800 for implementing the above method according to an embodiment of the present invention is described. It can be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, the readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0157] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0158] The computer-readable storage medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted by any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0159] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or, it can be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0160] This embodiment proposes a high-frequency-based infrared palm liveness detection method. The high-frequency part in the infrared palm image is extracted through Fourier transform, and the details such as blood vessels in the image are enhanced by filtering processing. Then, the high-frequency infrared palm image is reconstructed through inverse Fourier transform, and the liveness recognition model is used for recognition and judgment. It can not only effectively extract and utilize the high-frequency components in the infrared palm image, improve the accuracy and robustness of liveness detection, but also adapt to various lighting environments and camouflage attack scenarios, providing a strong guarantee for the security of the biometric recognition system.

[0161] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0162] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various deformations or modifications within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A high-frequency infrared palm liveness detection method, characterized in that: include: Step S1: Acquire infrared palm image; Step S2: Performing Fourier transform on the infrared palm image to obtain spectrum information and extract the high frequency part; Step S3: filtering the high frequency part to enhance the detail information such as blood vessels in the image; Step S4: performing inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image; Step S5: Performing liveness recognition on the high-frequency infrared palm image using a liveness recognition model.

2. The high-frequency infrared palm liveness detection method according to claim 1 is characterized in that: Step S2 includes: Step S21: performing a two-dimensional Fourier transform on the infrared palm image; Step S22: analyzing the result of the two-dimensional Fourier transform to obtain spectrum information; Step S23: extracting the high frequency part according to the spectrum information.

3. The high-frequency infrared palm liveness detection method according to claim 1 is characterized in that: Step S3 includes: Step S31: selecting a high-frequency filter according to the characteristics of the high-frequency components; Step S32: Apply the high frequency filter to the high frequency part.

4. The high-frequency infrared palm liveness detection method according to claim 1, characterized in that: Step S4 includes: Step S41: performing a two-dimensional inverse Fourier transform on the high frequency part; Step S42: reconstructing a high-frequency infrared palm image according to the result of the two-dimensional inverse Fourier transform.

5. The high-frequency infrared palm liveness detection method according to claim 1 is characterized in that: Step S5 includes: Step S51: extracting features from the high-frequency infrared palm image to obtain high-frequency palm features; Step S52: input the high-frequency palm features into a liveness recognition model to obtain a liveness recognition result.

6. The high-frequency infrared palm liveness detection method according to claim 1, characterized in that: The living body recognition model uses the infrared palm image as supervision during training.

7. The high-frequency infrared palm liveness detection method according to claim 1 is characterized in that: When the liveness recognition model is trained, different high-frequency parts are obtained by setting different parameters, thereby improving the adaptability of the liveness recognition model.

8. A high-frequency-based infrared palm liveness detection system, used to implement the high-frequency-based infrared palm liveness detection method according to any one of claims 1 to 7, characterized in that: include: An acquisition module, used for acquiring infrared palm images; A transformation module, used for performing Fourier transformation on the infrared palm image to obtain spectrum information and extract the high-frequency part; A filtering module, used for filtering the high frequency part to enhance the detail information such as blood vessels in the image; An inverse transform module, used for performing inverse Fourier transform on the high-frequency part to obtain a high-frequency infrared palm image; The recognition module is used to perform liveness recognition on the high-frequency infrared palm image using a liveness recognition model.

9. A high-frequency infrared palm liveness detection device, characterized in that: include: processor; a memory storing executable instructions of the processor; The processor is configured to execute the steps of the high-frequency-based infrared palm liveness detection method according to any one of claims 1 to 7 by executing the executable instructions.

10. A computer-readable storage medium for storing a program, characterized in that: When the program is executed, the steps of the high-frequency-based infrared palm liveness detection method described in any one of claims 1 to 7 are implemented.

Images