A two-image alignment method based on non-homologous double purposes

Abstract

A two-image alignment method based on non-homologous double purpose brush palm recognition, comprising the following steps: step S1: correcting the distortion of the original first image and the original second image respectively, and then performing polar line correction to obtain the corrected first image I ir and the corrected second image I rgb ; step S2: detecting the palm of the first image I ir and the second image I rgb respectively, and obtaining the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb respectively; step S3: extracting the palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and generating new images E ir and E rgb respectively; step S4: registering the palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and calculating the parallax d of the first image I ir and the second image I rgb ; step S5: calculating the depth of the palm edge and the depth of the palm center; step S6: aligning the first image I ir and the second image I rgb .

Description

TECHNICAL FIELD

[0001] The present application relates to the field of palm recognition camera, in particular, to a palm recognition two-image alignment method based on non-homologous binoculars. BACKGROUND

[0002] Since the palm lines of a human are more stable, palm recognition is a more stable and secure biometric technology than face recognition, and thus can be used to identify a person's identity through palm recognition, and further used in security checks, payments, identity recognition and other fields. Palm recognition is a technology with broad application prospects.

[0003] In the prior art, some palm recognition systems use a palm holder to fix the palm to determine the palm position, but this is inconvenient in actual application. Some palm recognition systems use p-sensors to estimate the palm depth, and then perform pixel-level alignment on the first image and the second image according to the p-sensors. In practice, generally, in order to measure the range and distance, it is necessary to take a complete image of the palm at a close distance and to clearly image the palm at a long distance. This requires a large FOV of the camera, and a large FOV of the camera will result in a small proportion of the palm in the image at a long distance. At this time, if the p-sensors are used for two-image alignment, a large number of p-sensors are required, which will interfere with each other and be difficult to design compatibly, and the cost will also be greatly increased.

[0004] The binocular system in the prior art uses homologous binoculars to obtain clear depth data of the target object, but the binocular system has a large amount of calculation and needs a separate system-on-chip to process the related data, thereby causing a series of problems such as transmission delay and recognition lag. SUMMARY

[0005] Therefore, the present application uses a non-homologous binocular system composed of a first camera and a second camera to directly process the first image and the second image, saves p-sensors, calculates the palm center disparity using edge depth and disparity, and aligns the first image and the second image, so that the palm recognition calculation is greatly reduced, the devices are less dependent, the device space is more easily saved, and the compatibility is better, the cost is reduced, and the promotion of palm recognition application is facilitated.

[0006] In a first aspect, the present application provides a palm recognition two-image alignment method based on non-homologous binoculars, characterized in that it comprises the following steps:

[0007] Step S1: Distortion correction is performed on the original first image and the original second image, and then epipolar correction is performed to obtain a corrected first image I ir and a corrected second image I rgb; wherein the original first image and the original second image are non-homologous images;

[0008] Step S2: detecting a palm in the first image I ir and the second image I rgb respectively, to obtain a palm region ROI ir of the first image I ir and a palm region ROI rgb of the second image I rgb respectively;

[0009] Step S3: extracting a palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and generating new images E ir and E rgb respectively;

[0010] Step S4: registering the palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and calculating a disparity d of the first image I ir and the second image I rgb ;

[0011] Step S5: calculating a depth of the palm edge and a depth of a palm center;

[0012] Step S6: aligning the first image I ir and the second image I rgb .

[0013] Optionally, the method for recognizing two-image alignment by brushing a palm based on non-homologous double purposes, wherein the step S3 comprises:

[0014] Step S31: performing image segmentation on the first image I ir and the second image I rgb according to the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and setting non-palm regions to zero;

[0015] Step S32: performing image segmentation on the first image I ir and the second image I irand the second image I rgb palm region ROI rgb are extracted, and new images E ir and E rgb are generated respectively.

[0016] Optionally, the two-image alignment method for palm recognition based on non-homologous double purposes, wherein the step S4 comprises:

[0017] Step S41: the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb are registered;

[0018] Step S42: the registration is verified, and the wrong matching is removed;

[0019] Step S43: the parallax d of the first image I ir and the second image I rgb is calculated.

[0020] Optionally, the two-image alignment method for palm recognition based on non-homologous double purposes, wherein the step S5 comprises:

[0021] Step S51: the depth of the palm edge is calculated according to the parallax d;

[0022] Step S52: the palm center is calculated according to the palm region of the first image I ir and the palm region of the second image I rgb .

[0023] Step S53: the average depth of the palm center is calculated according to the depth of the palm edge and the position of the palm center.

[0024] Optionally, the two-image alignment method for palm recognition based on non-homologous double purposes, wherein the step S6 comprises:

[0025] Step S61: the parallax b of the first image I ir and the second image I rgb at the palm center is calculated according to the depth of the palm center;

[0026] Step S62: the images are translated along the epipolar plane, so that the palm centers of the first image I ir and the second image I rgb are coincided.

[0027] Optionally, the two-image alignment method for palmprint recognition based on non-homologous double purposes has the feature that before the processing of the images in any step, the first image I ir , the second image I rgb , the new image E ir or the new image E rgb is compressed.

[0028] Optionally, the two-image alignment method for palmprint recognition based on non-homologous double purposes has the feature that the compression ratio of the first image I ir and the second image I rgb is different.

[0029] Optionally, the two-image alignment method for palmprint recognition based on non-homologous double purposes has the feature that the compression ratio of the new image E ir or the new image E rgb is different.

[0030] In the second aspect, the application provides a two-image alignment device for palmprint recognition based on non-homologous double purposes, which has the feature of comprising:

[0031] a processor;

[0032] a memory module in which executable instructions of the processor are stored;

[0033] wherein the processor is configured to execute the steps of the two-image alignment method for palmprint recognition based on non-homologous double purposes by executing the executable instructions.

[0034] In the third aspect, the application provides a computer-readable storage medium for storing a program, which has the feature that the program, when executed, implements the steps of the two-image alignment method for palmprint recognition based on non-homologous double purposes.

[0035] Compared with the prior art, the application has the following advantages:

[0036] The application uses the first image and the second image as original data, without the need for p-sensor devices, which reduces the input conditions for palmprint recognition, thereby simplifying the corresponding hardware devices, making them smaller and easier to integrate, and facilitating the miniaturization of the devices.

[0037] The image adopted by the present application can be shared with other palm recognition functions, so that one image can be used for multiple functions, thereby maximizing the function of one image, saving steps and device space. For example, when the first image is an infrared image, it can not only be used for alignment and reconstruction, but also for living body detection; when the second image is a color image, it can not only be used for alignment and reconstruction, but also for palmprint recognition. The present application aligns two different types of images to make palm detection, palm segmentation, edge extraction, edge matching, depth reconstruction and other steps more convenient, saving subsequent image processing steps and improving processing efficiency.

[0038] The present application uses a first camera and a second camera to image the palm, realizing pixel-level alignment of two images without depth map. This paper proposes using a first camera and a second camera to form a binocular, according to the binocular imaging theory to perform three-dimensional reconstruction on the edge of the palm, and then according to the depth information of the palm edge to realize pixel-level alignment of the first image and the second image.

[0039] The present application forms a binocular system with a first camera and a second camera that are not homologous, but due to their non-homology, the matching and other problems in the prior art cannot be effectively solved. The present application realizes matching and three-dimensional reconstruction of the non-homologous binocular system. For example, when the first camera is a near-infrared camera and the second camera is a color camera, when shooting the palm image, due to the absorption of infrared light by the veins, the veins will be darker; while the second image simultaneously shoots the palm image, mainly imaging the surface texture of the palm, which will result in greater difference between the first image and the second image, thereby making it impossible to effectively process using the prior art.

[0040] The present application does not need to process the entire palm image, only the palm edge in the image, greatly reducing the amount of data processed, thereby reducing the requirements for chips and the like, and no longer needing a separate system-on-chip for processing, which can be integrated into the chip provided by the camera, making the structure of the camera simpler and reducing costs, realizing low-cost palm brushing applications, which is conducive to commercial promotion.

[0041] The present application uses depth information to obtain the parallax of the palm center, and then performs translation, obtaining more accurate data and better translation effect, thereby achieving higher alignment quality, especially in the case of data loss and other situations in actual application, which can ensure the alignment accuracy of the data and has better adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort. Other features, objects and advantages of the present application will become more apparent through reading the following detailed description of the non-limiting embodiments with reference to the accompanying drawings:

[0043] Figure 1 A step flow chart of a two-image alignment method based on non-homologous double purposes of palm recognition in an embodiment of the present application;

[0044] Figure 2 A palm detection image in an embodiment of the present application;

[0045] Figure 3 A palm edge image in an embodiment of the present application;

[0046] Figure 4 A palm edge registration image in an embodiment of the present application;

[0047] Figure 5 A palm edge depth image in an embodiment of the present application;

[0048] Figure 6 A step flow chart of a new image acquisition method in an embodiment of the present application;

[0049] Figure 7 A step flow chart of a first image and a second image parallax calculation method in an embodiment of the present application;

[0050] Figure 8 A palm edge parallax image in an embodiment of the present application;

[0051] Figure 9 A step flow chart of a palm edge depth and palm center depth calculation method in an embodiment of the present application;

[0052] Figure 10 A step flow chart of a first image and a second image alignment method in an embodiment of the present application;

[0053] Figure 11 A set of palm alignment images in an embodiment of the present application;

[0054] Figure 12 A structure schematic diagram of a two-image alignment device based on non-homologous double purposes of palm recognition in an embodiment of the present application;

[0055] Figure 13 A computer readable storage medium schematic diagram in an embodiment of the present application. Detailed Implementation

[0056] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention. These all fall within the scope of protection of the present invention.

[0057] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0058] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0059] This invention provides a method for aligning two images based on non-homogeneous binocular palm recognition, which aims to solve the problems existing in the prior art.

[0060] The technical solutions of the present invention and how they solve the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments 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 now be described with reference to the accompanying drawings.

[0061] Figure 1 This is a flowchart illustrating the steps of a method for aligning two images based on non-homogeneous binocular palm print recognition, as described in an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a method for aligning two images based on non-homogeneous binocular palm print recognition, comprising the following steps:

[0062] Step S1: Correct the distortion of the original first image and the original second image respectively, and then perform epipolar correction to obtain the corrected first image I. ir and the corrected second image I rgb .

[0063] In this step, the original first image and the original second image are non-homologous images, i.e. images obtained by different technologies. The original first image is subjected to distortion correction and then polar correction to obtain a corrected first image I ir . The original second image is subjected to distortion correction and then polar correction to obtain a corrected second image I rgb . Since distortion is caused by the principle of lens imaging, the distortion correction of the original first image and the original second image needs to be corrected according to the respective parameters of the acquisition device. Polar correction is a correction for a binocular system, which is to rotate the two cameras and redefine a new image plane, so that the polar lines are collinear and parallel to a certain coordinate axis (usually the horizontal axis), which simultaneously establishes a new stereo pair. After correction, the same matching point pair is located in the same row of the two views, which means that they only have a difference in horizontal coordinates (or column coordinates), and this difference is called parallax. However, since the images used are the first image and the second image, there are differences in the content they capture, and it is not possible to directly use current technology to solve the parallax. When the first camera captures the palm image, the veins have a certain absorption effect on infrared light, causing the veins to be relatively dark; while the second camera captures the palm image, it mainly images the surface texture of the palm, and the imaging of the two palm textures is difficult to match directly. At the same time, the palms of people of different body types differ greatly, making the difference between the infrared image and the second image even greater, making it difficult to effectively match. This step corrects the two images to make the data more accurate, thereby making the subsequent matching more accurate. It should be noted that the original first image and the original second image used in this embodiment are usually obtained by a calibrated binocular system, and one of the binocular systems is a first camera and the other is a second camera. Among them, the first camera is used to acquire the first image, the second camera is used to acquire the second image, and the first camera and the second camera acquire the target image at the same time. For example, the first camera is a near-infrared camera, the first image is a near-infrared image, the second camera is a color camera, and the second image is a color image.

[0064] This step allows images with large distortion to be better processed, so that when the palm is close to the shooting device, i.e. when the FOV is large, accurate results can also be obtained, thereby improving the effective recognition distance range of the palm.

[0065] Step S2: detecting the palm in the first image I ir and the second image I rgb respectively, to obtain the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb.

[0066] In this step, a palm detection model is used to detect the first image I ir and the second image I rgb respectively, so as to determine whether a palm exists. If a palm exists, the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb are obtained. As shown in FIG. 3, the palm region ROI ir and the palm region ROI rgb are both the smallest rectangles containing the palm, that is, the four edges of the rectangle are tangent to the edges of the palm. The palm detection model used in this step can be any model that can achieve the function of palm detection, which is not limited in the present embodiment. Figure 2

[0067] Step S3: The palm edges of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb are extracted, and new images E ir and E rgb are generated respectively.

[0068] In this step, the edges of the palm are extracted according to an edge extraction algorithm, and new images E ir and E rgb are generated. The edge extraction algorithm can be implemented in various ways, such as an algorithm based on a designed edge extraction operator (convolution template), including but not limited to sobel, prewit, robert, LoG, etc. The edge extraction algorithm can also be obtained by using adaptive algorithms or machine learning trained models, which are not limited in the present embodiment. As shown in FIG. 7, the recognized palm edges have a certain width, and because the captured object is the same palm, the edges also have very good consistency compared to the inside of the palm, which can achieve better processing results. Figure 3

[0069] Step S4: The palm edges of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb are registered, and the parallax d of the first image I ir and the second image I rgb is calculated.

[0070] ​​In this step, in the new image E ir and E rgb The first image I is obtained by registering the image using the edge of the palm. ir and the second image I rgb The parallax d. For example... Figure 4 As shown, when performing palm edge registration, all edge points can be matched, or only some points can be matched. Figure 4 The method uses 68 annotation points for registration, achieving a balance between effective palm representation and computational load. Ideally, the number of annotation points used for palm edge registration should be between 39 and 136. If the number is less than 39, the palm shape cannot be adequately represented; while if the number is greater than 136, the computational load is too high, consuming excessive computational resources, but the improvement in recognition performance is very limited.

[0071] Step S5: Calculate the depth of the edge of the palm and the depth of the center of the palm.

[0072] In this step, after calculating the first image I... ir and the second image I rgb After determining the parallax 'd', the depth of the palm edge and the depth of the palm center can be calculated using the binocular principle. The palm center is the point where the differences in distance from the palm to the left and right sides, as well as the distance to each base of the palm, are minimal. For example... Figure 5 As shown, the edge of the palm is the edge of three-dimensional space. When projected onto a two-dimensional plane, it still appears as a line with a certain width. The palm edge depth map only shows the depth data of the palm edge; the data for other parts is 0. The depth of the palm center is obtained using the same method. The depth of the palm center is obtained by calculating the center point of the palm.

[0073] Step S6: For the first image I ir and the second image I rgb Alignment.

[0074] In this step, the first image I is processed based on the data obtained in the preceding steps. ir Second image I rgb Alignment. During alignment, data obtained from any one or more of the aforementioned steps can be processed, such as aligning by the edge of the palm, aligning by the center of the palm, or aligning by both the edge and center of the palm. When translating the image, only the first image I can be translated. ir Alternatively, you can simply translate the second image I. rgb It can also be used for the first image I ir Second image I rgb Translate them all so that the centers of the palms coincide.

[0075] In some embodiments, the first image I ir , the second image I rgb , the new image E ir or the new image E rgb is compressed before the processing of the image in any step is performed. The compression ratio of the first image I ir and the second image I rgb is different. The compression ratio of the new image E ir or the new image E rgb is different. For example, the first image I ir and the second image I rgb are compressed before step S2 is performed, and the compression ratio is 4 and 2 respectively, so that the size of the image is closer, and the accuracy of the processing can be ensured. Preferably, the closer the palm is to the camera, the larger the compression ratio; the farther the palm is to the camera, the smaller the compression ratio. When the palm occupies 70% of the area of the image, the compression ratio of the first image is not less than 5; when the palm occupies 50% of the area of the image, the compression ratio of the first image is not less than 2. The compression ratio of the first image is more than 3 times the compression ratio of the second image.

[0076] Figure 6 is a flow chart of a step of obtaining a new image in an embodiment of the present application. As shown in Figure 6 , unlike the foregoing embodiments, the method of obtaining a new image in an embodiment of the present application comprises the following steps:

[0077] Step S31: the first image I ir and the second image I rgb are respectively subjected to image segmentation according to the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb , and the non-palm region is set to zero.

[0078] In this step, in addition to image segmentation, the non-palm region is set to zero to improve the extraction effect of the subsequent edge extraction algorithm and to be able to cope with more complex scenes. When the palm wears rings, hand ornaments and other types of decorations, the scene will be more complex, so setting the non-palm region to zero can make the subsequent processing easier.

[0079] Step S32: the palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb is extracted on the image obtained in step S31, and a new image Eir and E rgb .

[0080] In this step, only the palm region is processed, and the palm edge is extracted, so that the processed region is smaller, and the processing speed is faster. The new image E ir and E rgb only contains the palm edge, and can be used for processing in subsequent steps, thereby greatly reducing the processing amount and improving efficiency.

[0081] Figure 7 is a flowchart of a step of calculating the disparity of the first image and the second image in an embodiment of the present application. As shown in Figure 7 , unlike the foregoing embodiments, the method for calculating the disparity of the first image and the second image in an embodiment of the present application comprises the following steps:

[0082] Step S41: registering the palm edge of the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image I rgb .

[0083] In this step, the palm edges in the two images are registered first. According to prior knowledge, the information of ROI ir and ROI rgb , the edges are registered using a traditional method or deep learning. Preferably, in the registration, a feature-based method is adopted, and the labeled points in the foregoing embodiments are feature points. Through registration, the registration of the two images can be realized.

[0084] Step S42: verifying the registration and eliminating incorrect matches.

[0085] In this step, according to prior knowledge, incorrect matches are eliminated; for example, the disparity needs to have consistency, topological relationship, etc. Compared with the matching scheme in the prior art, the two images matched in this embodiment are obtained by different technologies, so there will be incorrect matches due to different technical characteristics, but these usually have certain regularity. When eliminating incorrect matches, corresponding filtering models can also be used according to different technical characteristics.

[0086] Step S43: calculating the disparity d of the first image I ir and the second image I rgb .

[0087] In this step, the palm region ROI ir of the first image I ir and the palm region ROI rgb of the second image Irgb the palm edge of the first image I ir and the second image I rgb d. As shown in the figure, the disparity d is obtained by calculating x1 and x2. Figure 8

[0088] The embodiment calculates the disparity of the two images by registering the palm edge of the palm region and eliminating the false matching, so as to improve the quality of registration, make the calculation of disparity more accurate, and make the subsequent data more accurate.

[0089] Figure 9 The figure is a flow chart of a method for calculating the depth of the palm edge and the depth of the palm center in the embodiment of the application. As shown in the figure, different from the foregoing embodiment, the method for calculating the depth of the palm edge and the depth of the palm center in the embodiment of the application comprises the following steps: Figure 9

[0090] Step S51: calculating the depth of the palm edge according to the disparity d.

[0091] In this step, the depth of the palm edge is calculated by using the binocular principle in combination with the disparity d.

[0092] Step S52: calculating the palm center according to the palm region of the first image I ir and the palm region of the second image I rgb

[0093] In this step, the center point is determined according to the shape of the palm region, that is, the palm center. Since the palms on the two images have been registered, the center point on any image can be calculated to obtain the palm center.

[0094] Step S53: calculating the average depth of the palm center according to the depth of the palm edge and the position of the palm center.

[0095] ​​​In this step, the average depth of the palm center is calculated according to the depth of the palm edge and the position of the palm center, and the weight is higher when the distance to the palm center is closer. It should be noted that when calculating the average depth of the palm center, only the depth of the palm edge and the position of the palm center are needed, and the depth value inside the palm is not needed, which is different from the prior art. Due to the irregularity of the shape of the palm, the distance from each point on the palm edge to the center point is not the same, so different weight values are used for calculation according to the distance from each point to the palm center. The distribution of weight values is not fixed, but can be adjusted according to the position of the palm center in the palm and the posture of the palm. The average depth of the palm center obtained by the embodiment is different from the actual depth value of the palm center, but this does not affect the alignment operation of the palm, on the contrary, it makes the subsequent alignment operation simpler and more accurate.

[0096] In this embodiment, the average depth of the palm center is calculated by the depth of the palm edge and the position of the palm center, so that the palm inside does not need to be calculated, that is, the depth value of the palm center can be obtained, avoiding the problem of difficult matching and calculation inside the palm. While ensuring the accuracy of the data, the amount of data calculation is smaller, and the efficiency is improved.

[0097] Figure 10 A step flow chart for aligning a first image and a second image in an embodiment of the present application is shown in FIG. 1. Figure 10 As shown in FIG. 1, unlike the previous embodiments, the method for aligning a first image and a second image in an embodiment of the present application comprises the following steps:

[0098] Step S61: calculating the depth of the palm center of the first image I ir and the second image I rgb at the palm center.

[0099] In this step, the disparity b at the depth of the palm center is obtained. The calculation method of the disparity b of the palm center is the same as that of the disparity d of the palm edge.

[0100] Step S62: translating the images along the epipolar plane to make the palm centers of the first image I ir and the second image I rgb coincide.

[0101] In this step, the images are translated using the epipolar plane in the binocular system, specifically, the images are translated using the epipolar plane where the palm center is located, so that the palm centers of the first image I ir and the second image I rgb coincide, thereby completing the alignment operation. Compared with the previous part of the embodiment, the new image E ir and E rgbIn operation, the embodiment directly processes the first image I ir and the second image I rgb In operation, the two images are processed so as to complete the alignment operation. As shown in Figure 11 The aligned images are more consistent, so that the palm can be better recognized and processed.

[0102] The embodiment utilizes the epipolar plane to translate the images, so that the movement of the images is more controllable, which is beneficial to the consistency of the operation and ensures the quality of the obtained aligned images.

[0103] The embodiment also provides a two-image alignment device for palm recognition based on non-homologous binoculars, which comprises a processor and a memory having executable instructions of the processor stored therein. The processor is configured to execute the steps of the two-image alignment method for palm recognition based on non-homologous binoculars.

[0104] As described above, the embodiment uses the depth camera of the binocular system composed of the first camera and the second camera to obtain the first image and the second image, and aligns the two different types of images by the method in the foregoing embodiment, so as to overcome the differences between the different types of images and achieve the purpose of stable and rapid alignment.

[0105] Those skilled in the art can understand that each aspect of the present application can be implemented as a system, a method or a program product. Therefore, each aspect of the present application can be specifically implemented as a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software, which can be collectively referred to as "circuitry", "module" or "platform" here.

[0106] Figure 12 is a structural schematic diagram of a two-image alignment device for palm recognition based on non-homologous binoculars in the embodiment of the present application. The electronic device 600 according to this embodiment of the present application will be described below with reference to Figure 12 Figure 12 The displayed electronic device 600 is only an example and should not limit the functions and use range of the embodiment of the present application.

[0107] As shown in Figure 12 , the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 can 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.

[0108] ​The storage unit stores program codes which can be executed by the processing unit 610, so that the processing unit 610 performs the steps of the two figure alignment method based on non-homologous double-strand ends described in the above description according to various exemplary embodiments of the present application. For example, the processing unit 610 can perform the steps as shown in FIG. 6. Figure 1

[0109] The storage unit 620 can include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 6201 and / or a cache memory 6202, and can further include a read-only memory (ROM) 6203.

[0110] The storage unit 620 can further include a program / utility 6204 having a set of programs / modules 6205, including an operating system, one or more application programs, other programs, and programmatic data, each or some combination thereof, which can include implementation of a network environment.

[0111] The bus 630 can represent one or more of several types of bus structures, including a storage unit bus or bus controller, a peripheral bus, a graphics acceleration port, a processing unit bus, or a local bus using any of a variety of bus architectures.

[0112] The electronic device 600 can also communicate with one or more external devices 700 such as a keyboard or pointing device, a Bluetooth device, etc.; other devices that enable a user to interact with the electronic device 600; and / or one or more devices that enable the electronic device 600 to communicate with one or more other computing devices. Such communication can occur via an input / output (I / O) interface 650. Still yet, the electronic device 600 can communicate with one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or the Internet, through a network adapter 660. The network adapter 660 can be any of a variety of modems, including cable modems, telephone modem, and fiber optic modems. The network adapter 660 can be communicatively coupled to the other components of the electronic device 600 via a bus 630. It should be appreciated that the bus 630 can be one of a variety of bus architectures, including, for example, a memory bus, a video bus, a peripheral bus, a local bus, and any combinations thereof. Figure 12 It should be appreciated that the electronic device 600 can also employ other hardware and / or software modules that are not shown in FIG. 6, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0113] ​The embodiment of the present application also provides a computer readable storage medium for storing a program, which, when executed, implements the steps of the two-image alignment method based on non-homologous double purposes for palm recognition. In some possible implementation manners, various aspects of the present application can also be implemented in the form of a program product, which includes program codes for causing a terminal device to perform the steps described in the above two-image alignment method based on non-homologous double purposes for palm recognition according to various exemplary embodiments of the present application when the program product is run on the terminal device.

[0114] As shown above, the program of the computer readable storage medium of the embodiment, when executed, acquires the first image and the second image by using the depth camera of the binocular system composed of the first camera and the second camera, and aligns the two different types of images by the method in the above embodiment, so as to overcome the difference between the different types of images and achieve the purpose of stable and fast alignment.

[0115] Figure 13 is a structural schematic diagram of the computer readable storage medium in the embodiment of the present application. Referring to Figure 13 As shown in the above, the program product 800 for implementing the above method according to the embodiment of the present application can adopt a portable compact disc read-only memory (CD-ROM) and includes program codes, and can be run on a terminal device, such as a personal computer. However, the program product of the present application is not limited to this, and in the present document, the readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or apparatus.

[0116] The program product can adopt any combination of one or more readable media. The readable medium 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, device or apparatus, 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 disc, 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.

[0117] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0118] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone 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 cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type 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., via the Internet using an Internet service provider).

[0119] In this embodiment of the invention, a first image and a second image are acquired using a depth camera of a binocular system consisting of a first camera and a second camera. The two different types of images are aligned using the method described in the foregoing embodiment, overcoming the differences between the different types of images and achieving stable and fast alignment.

[0120] The various embodiments described in this specification are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0121] The specific embodiments of the present application are described above. It needs to be understood that the present application is not limited to the specific embodiments described above, and various modifications or changes can be made by those skilled in the art within the scope of the claims, which do not affect the essence of the present application.

Claims

1. A method for recognizing two figure alignment based on non-homologous double purposes of brush palm, characterized by, comprising the steps of: Step S1: respectively performing distortion correction and epipolar correction on the original first image and the original second image to obtain a corrected first image and a corrected second image ; wherein the original first image and the original second image are non-homologous images ; wherein the original first image and the original second image are non-homologous images Step S2: For the first image and the second image The first image is obtained by detecting the palm of the hand separately. palm area and the second image palm area ; Step S3: For the first image palm area and the second image palm area The edges of the palm are extracted and new images are generated separately. and ; Step S4: For the first image palm area and the second image palm area The palm edge is registered, and the first image is calculated. and the second image The parallax d; Step S5: calculating the depth of the palm edge and the depth of the palm center; Step S6: aligning said first image and said second image to each other; said step S5 comprising: Step S51: calculating the depth of the palm edge according to the parallax d; Step S52: calculating the palm center according to the palm region of the first image and the palm region of the second image Step S53: calculating the palm center according to the palm region of the first image Step S53: calculating the average depth of the palm center according to the depth of the palm edge and the position of the palm center; said step S6 comprising: Step S61: calculating the first image according to the palm center depth and the second image the parallax b at the palm center; Step S62: Translate the image along the epipolar plane so that the palm centers of the first image and the second image coincide.

2. The two figure alignment method based on non-homologous double palms according to claim 1, characterized in that, said step S3 comprising: Step S31: image segmentation is performed on the first image and the second image respectively according to the palm region of the first image and the palm region of the second image and the non-palm region is set to zero; Step S32: On the image obtained in step S31, apply the first image... palm area and the second image palm area The edges of the palm are extracted and new images are generated separately. and .

3. The two figure alignment method based on non-homologous double purposes of brush palm identification according to claim 1, characterized in that, said step S4 comprising: Step S41: For the first image palm area and the second image palm area Register the palm edge; Step S42: verifying the registration and eliminating the false matches; Step S43: calculating the disparity d of the first image and the second image .

4. The two figure alignment method based on non-homologous double palms according to claim 1, characterized in that, The first image , the second image , the new image or the new image is compressed before the processing of the image in performing any of the steps.

5. The two figure alignment method based on non-homologous double palms according to claim 4, characterized in that, The compression ratio of the first image and the second image is different.

6. The two figure alignment method based on non-homologous double palms according to claim 4, characterized in that, to the new image or the compression ratio of the new image is different.

7. A method for recognizing two figure alignment based on non-homologous double purposes, characterized in that, comprising the steps of: Step S1: respectively performing distortion correction and epipolar correction on the original first image and the original second image to obtain a corrected first image and a corrected second image ; wherein the original first image and the original second image are non-homologous images ; wherein the original first image and the original second image are non-homologous images Step S2: For the first image and the second image The first image is obtained by detecting the palm separately. palm area and the second image palm area ; Step S3: extracting the palm region of the first image and the palm region of the second image and generating new images and respectively and ; Step S4: For the first image palm area and the second image palm area The palm edge is registered, and the first image is calculated. and the second image The parallax d; Step S5: calculating the depth of the palm edge and the depth of the palm center; Step S6: aligning said first image and said second image to each other; said step S6 comprising: Step S61: calculating the first image according to the palm center depth and the second image the parallax b at the palm center; Step S62: Translate the image along the epipolar plane so that the palm centers of the first image and the second image coincide.

8. A device for recognizing two figure alignment based on non-homologous double purposes, characterized in that, comprising: a processor; a memory module having executable instructions of the processor stored therein; wherein the processor is configured to execute the executable instructions to perform the steps of the two-image alignment method based on the non-homologous double purpose palm recognition according to any one of claims 1 to 7.

9. A computer readable storage medium for storing a program, characterized in that, The program, when executed, implements the steps of the two-image alignment method based on the non-homologous double purpose palm recognition according to any one of claims 1 to 7.

Images