Brush palm binocular camera and electronic equipment

Abstract

The invention discloses a brush palm binocular camera and electronic equipment, and the camera is characterized in that the camera comprises a first receiver which is used for receiving a first signal of a palm, and generating a first image; the second receiver is used for receiving a second signal of the palm and generating a second image; the processor is used for detecting a first hand area on the first image, detecting a second hand area on the second image, aligning the first hand area with the second hand area, zooming the first hand area and the second hand area to a preset size at the same magnification, and outputting the first hand area and the second hand area; and depth calculation is carried out on the first hand region and the second hand region. According to the method, the calculation amount of search and reconstruction is greatly reduced, a special Asic chip or special hardware is not needed, the depth of the hand is calculated in real time, and help is provided for subsequent gesture recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of binocular cameras, and in particular to a palm-swiping binocular camera and electronic equipment. Background Art

[0002] In gesture recognition, depth information is very beneficial for improving the performance of gesture recognition. It can provide more spatial context, reduce the ambiguity of two-dimensional images, and improve the robustness of the algorithm under various environmental conditions.

[0003] The main functions of depth information are:

[0004] 1. Added spatial information: Depth data adds an extra dimension to gestures, helping the algorithm understand the shape and movement of gestures in space.

[0005] 2. Improve robustness: In environments with complex backgrounds or changing lighting, depth information can help the algorithm better distinguish between hands and background.

[0006] 3. Reduce ambiguity: When only two-dimensional information is available, some gestures may look similar. Depth data can help resolve this visual ambiguity and improve the accuracy of gesture recognition.

[0007] 4. Handling occlusion: Self-occlusion, which is common in gestures (e.g. one finger occluding another finger), is easier to distinguish and handle in 3D space.

[0008] Binocular cameras can obtain dense depth through binocular stereo matching. Unlike tof or monocular structured light depth cameras, binocular cameras do not require special hardware except for the two cameras, so their hardware cost is relatively low, and they have good autonomy and flexibility, so they are widely applicable. However, their disadvantages are also obvious. The binocular stereo matching algorithm is computationally intensive and requires a lot of computing resources. Therefore, in actual use, dedicated Asic chips or dedicated hardware are generally required for depth calculation, which increases the cost.

[0009] The disclosure of the above background technology content is only used to assist in understanding the inventive concept and technical solution of the present invention. It does not necessarily belong to the prior art of this patent application. In the absence of clear evidence that the above content has been disclosed on the filing date of this patent application, the above background technology should not be used to evaluate the novelty and creativity of the present application. Summary of the invention

[0010] To this end, the present invention identifies the hand area, scales the hand area, and obtains the depth value of the hand area by calculating the parallax, which greatly reduces the amount of calculation for search and reconstruction. There is no need to use a dedicated Asic chip or dedicated hardware, and the depth of the hand is calculated in real time to provide assistance for subsequent gesture recognition.

[0011] In a first aspect, the present invention provides a palm-swiping binocular camera, characterized in that it comprises:

[0012] A first receiver, used for receiving a first signal of a palm and generating a first image;

[0013] A second receiver, used for receiving a second signal of the palm and generating a second image;

[0014] A processor is used to detect a first hand area on the first image, detect a second hand area on the second image, align the first hand area with the second hand area, scale the first hand area and the second hand area to a preset size at the same magnification, and then perform depth calculation on the first hand area and the second hand area.

[0015] Optionally, the palm-scanning binocular camera is characterized by further comprising:

[0016] A light source is used to illuminate the palm according to the brightness of the first image.

[0017] Optionally, the palm-scanning binocular camera is characterized in that the processor comprises:

[0018] A detection module, configured to detect a first hand region on the first image and a second hand region on the second image;

[0019] an alignment module, configured to align the first hand region with the second hand region;

[0020] A scaling module, used for scaling the first hand area and the second hand area to a preset size at the same magnification;

[0021] A depth calculation module is used to perform depth calculation on the first hand area and the second hand area.

[0022] Optionally, the palm-scanning binocular camera is characterized in that the alignment module comprises:

[0023] an epipolar constraint unit, configured to perform epipolar constraint processing on the first hand region and the second hand region in the y-axis direction;

[0024] The alignment unit is used to align the first hand area and the second hand area to have the same pixel size range with the center point in the x-axis direction.

[0025] Optionally, the palm-scanning binocular camera is characterized in that the depth calculation module comprises:

[0026] A palm recognition unit, configured to obtain a first palm region in the first hand region and obtain a second palm region in the second hand region;

[0027] A first disparity unit, configured to estimate a search range according to a size of the first palm area or the second palm area, and calculate a disparity between the first palm area and the second palm area;

[0028] The first depth unit is used to calculate the depth value of each pixel in the first palm area according to the disparity and the parameters of the binocular camera.

[0029] Optionally, the palm-scanning binocular camera is characterized in that the depth calculation module comprises:

[0030] A finger recognition unit, used to obtain a first finger region in the first hand region and a second finger region in the second hand region;

[0031] A second disparity unit, used to determine the number of each finger in the first finger area and the second finger area, and calculate the disparity for fingers with the same number;

[0032] The second depth unit is used to calculate the depth value of each pixel in the first finger area according to the disparity and the parameters of the binocular camera.

[0033] Optionally, the palm-scanning binocular camera is characterized in that the second parallax unit calculates the search range of each finger according to the width of each finger, and then performs matching to calculate the parallax.

[0034] Optionally, the palm-scanning binocular camera is characterized in that the second parallax unit calculates the search range according to the maximum width of the finger, and then performs matching to calculate the parallax.

[0035] Optionally, the palm-scanning binocular camera is characterized in that the detection module comprises:

[0036] a detection unit, configured to detect a third hand region and a fourth hand region on the first image, and to detect a fifth hand region and a sixth hand region on the second image;

[0037] A pairing unit is used to pair the third hand region, the fourth hand region, the fifth hand region and the sixth hand region according to the epipolar constraint.

[0038] In a second aspect, the present invention provides an electronic device, characterized in that it includes any one of the palm-swiping binocular cameras described above.

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

[0040] The present invention first identifies the hand area, and in subsequent processing only processes the hand area, thereby reducing the amount of data processing and improving processing efficiency.

[0041] The present invention scales the image before calculation, thereby making the amount of calculation constant and avoiding the problem of exceeding the calculation capacity. At the same time, data conversion is performed when calculating the depth, so that the amount of data to be calculated is greatly reduced.

[0042] The present invention optimizes the hand area and scales it to greatly reduce the amount of calculation. The depth value of the hand area is obtained by calculating the parallax, which greatly reduces the amount of calculation for search and reconstruction. There is no need to use a dedicated Asic chip or dedicated hardware. The depth of the hand is calculated in real time, which helps subsequent gesture recognition and is cost-effective.

[0043] The present invention uses binocular technology to calculate the depth of the palm, and optimizes the amount of calculation, so that the amount of calculation is greatly reduced while the calculation accuracy can be guaranteed, and better depth data can be obtained than TOF technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings in the following descriptions are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work. By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, purposes and advantages of the present invention will become more obvious:

[0045] Figure 1 This is a schematic diagram of the structure of a palm-scanning binocular camera according to an embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of the structure of another palm-scanning binocular camera according to an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of the structure of a processor in an embodiment of the present invention;

[0048] Figure 4 This is a schematic diagram of a palm in an embodiment of the present invention;

[0049] Figure 5 is a schematic structural diagram of an alignment module in an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of the structure of a depth calculation module in an embodiment of the present invention;

[0051] Figure 7 A schematic diagram of palm area division in an embodiment of the present invention;

[0052] Figure 8 is a schematic diagram of the structure of another depth calculation module in an embodiment of the present invention;

[0053] Fig. 9 This is a schematic diagram of the structure of a detection module in an embodiment of the present invention;

[0054] Fig.10 The figure is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0055] 1--Hand area;

[0056] 2--Palm area;

[0057] 3--Finger area; DETAILED DESCRIPTION

[0058] The present invention is 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 are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several variations and improvements may be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0059] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein, for example, can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0060] For a stereo camera without a palm rest, the size of the hand in the image changes as the distance between the hand and the stereo camera changes. To ensure that the hand can be effectively recognized in a larger range, the resolution of the stereo camera is usually higher. When the hand is close to the stereo camera, the hand occupies a larger proportion in the image, and even if only the depth calculation is performed on the hand area, it still requires a large amount of calculation.

[0061] A palm-scanning binocular camera provided in an embodiment of the present invention is intended to be suitable for binocular depth calculation of hands at different distances, reduce the amount of calculation, and lower the requirements for chip computing power when the binocular camera is used for hand detection, thereby solving the problems existing in the prior art.

[0062] The following specific embodiments are used to describe in detail the technical solutions of the present invention and how the technical solutions of the present application solve the above-mentioned technical problems. The following 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 be described below in conjunction with the accompanying drawings.

[0063] The present invention identifies the hand area, scales the hand area, and obtains the depth value of the hand area by calculating the parallax, which greatly reduces the amount of calculation for search and reconstruction. It does not need to use a dedicated Asic chip or dedicated hardware, and calculates the depth of the hand in real time, providing assistance for subsequent gesture recognition.

[0064] Figure 1 FIG. 1 is a schematic diagram of the structure of a palm scanning binocular camera according to an embodiment of the present invention. Figure 1 As shown, a palm scanning binocular camera in an embodiment of the present invention includes:

[0065] The first receiver is used to receive a first signal from a palm and generate a first image.

[0066] Specifically, the first receiver is responsible for capturing the first signal of the palm, which generally refers to the image information of the palm obtained from a specific angle or position. The first receiver includes a camera sensor, such as a CMOS or CCD sensor, for converting the optical signal into an electrical signal to generate a digital image. The first image can be an RGB image, an infrared image, or any other type of image.

[0067] The second receiver is used to receive a second signal from the palm and generate a second image.

[0068] Specifically, similar to the first receiver, the second receiver captures a second signal of the palm from another angle or position to generate another image. The second receiver also includes a camera sensor for capturing images of the palm from different viewing angles. The second image is of the same type as the first image, such as both are RGB images, both are infrared images, etc.

[0069] A processor is used to detect a first hand area on the first image, detect a second hand area on the second image, align the first hand area with the second hand area, scale the first hand area and the second hand area to a preset size at the same magnification, and then perform depth calculation on the first hand area and the second hand area.

[0070] Specifically, the processor is the brain of the entire system, responsible for performing a variety of image processing tasks, including detecting the hand area, aligning the hand areas in the two images, scaling to a preset size, and performing depth calculations. The processor has built-in advanced image processing algorithms that can automatically detect the hand area and align and scale the hand areas captured by the two receivers. These algorithms need to be highly robust and accurate to cope with challenges such as different lighting conditions, palm postures, and angle changes. Based on the principle of binocular imaging, the processor can use the image information of the hand area captured by the two receivers to obtain the depth information of the palm by calculating the parallax. This depth information is crucial for subsequent applications such as palm recognition and security verification.

[0071] Figure 2 FIG. 1 is a schematic diagram of the structure of another palm scanning binocular camera according to an embodiment of the present invention. Figure 2 As shown, compared with the above-mentioned embodiment, another palm scanning binocular camera in the embodiment of the present invention further includes:

[0072] A light source is used to illuminate the palm according to the brightness of the first image.

[0073] Specifically, the role of the light source in the palm-swiping binocular camera is mainly to adjust its illumination intensity according to the brightness of the first image, to ensure that the acquired palm image is clear and visible, and to facilitate subsequent image processing and depth calculation.

[0074] Lighting plays a vital role in binocular camera systems, especially in high-precision applications such as palm scanning binocular cameras. Since such camera systems rely on the hand area in two images for depth calculation and 3D reconstruction, it is crucial to obtain clear and reliable images. Lighting helps improve image quality in low-light or high-light environments by providing stable and controllable lighting conditions, ensuring that the processor can accurately detect and align the hand area.

[0075] By controlling the intensity and spectral distribution of the light source, the visibility of biometric features such as palm texture and veins can be optimized, thereby improving the accuracy and robustness of recognition. The palm-swiping binocular camera system can achieve stable and reliable palm recognition functions under a wider range of environmental conditions. This is of great significance for improving the security and convenience of application scenarios such as access control systems and payment verification.

[0076] Figure 3 FIG. 1 is a schematic diagram of the structure of a processor in an embodiment of the present invention. Figure 3 As shown, a processor in an embodiment of the present invention includes:

[0077] A detection module is used to detect a first hand area on the first image and a second hand area on the second image.

[0078] Specifically, the detection module needs to perform two palm detections, namely, one palm detection for the first image and one palm detection for the second image. It is necessary to use a trained hand detection model to identify the hand area in the first image and the second image respectively. During processing, the first and second images are first preprocessed, such as denoising, contrast enhancement, etc., to facilitate subsequent hand detection. Then, image processing techniques (such as skin color detection, edge detection, contour recognition, or deep learning-based methods such as convolutional neural networks) are used to locate and extract the hand area. The hand area is usually the area in the image that matches the shape and size of the hand, which is determined by identifying specific features (such as skin color, finger shape, etc.). Once the hand area is detected, these areas are marked with bounding boxes or other shapes for subsequent processing. Figure 4 As shown, the hand region is often a part of the image. The subsequent processing of this embodiment is all performed on the hand region, and there is no need to process other parts, thereby reducing the amount of calculation.

[0079] An alignment module is used to align the first hand area with the second hand area.

[0080] Specifically, due to the different shooting angles of the two cameras, the position and shape of the hand in the two images will be different. Therefore, we need to align the two hand regions through image registration technology for subsequent depth calculation.

[0081] During alignment, first, feature points (such as corner points, edge points, etc.) are extracted from the two hand regions respectively, and then the correspondence between the two sets of feature points is found using a feature matching algorithm (such as SIFT, SURF, ORB, etc.). Secondly, based on the correspondence between the feature points, a mathematical method (such as the RANSAC algorithm) is used to estimate the geometric transformation (such as translation, rotation, scaling, etc.) between the two hand regions, thereby calculating the transformation matrix from the first hand region to the second hand region. Finally, the calculated transformation matrix is ​​used to transform the first hand region to a position aligned with the second hand region.

[0082] The scaling module is used to scale the first hand area and the second hand area to a preset size at the same magnification.

[0083] Specifically, the first hand area and the second hand area already completely contain the hand information, so scaling the first hand area and the second hand area will not result in the loss of hand information. Scaling the first hand area and the second hand area at the same magnification can ensure that the relative sizes of the first hand area and the second hand area remain unchanged, thereby ensuring the accuracy of the depth calculation. The preset size is usually a smaller size to achieve the purpose of reducing the amount of calculation. The setting of the preset size needs to take into account both the amount of calculation and the clarity. A preset size that is too large has limited effect on reducing the amount of calculation, while a preset size that is too small will cause the image to lose too much precision, and the final depth value is too small.

[0084] A depth calculation module is used to perform depth calculation on the first hand area and the second hand area.

[0085] Specifically, the depth information of the hand is calculated using the principle of binocular stereo vision. The depth calculation module performs depth calculation only for the first hand area and the second hand area, so that the reconstruction area can be very small. Since the binocular camera system has a fixed baseline distance and the two cameras shoot the same scene, the depth can be estimated by comparing the position difference (i.e., parallax) of corresponding points in the two hand areas. For each point (or each pixel) in the first hand area, find the corresponding point in the second hand area and calculate the parallax between them. Then, using the known baseline distance, focal length and other parameters and parallax, the depth value of the point is calculated by the triangulation principle. Repeat this process to generate a depth map of the entire hand area. The obtained depth map can be used for a variety of applications, such as gesture recognition, virtual reality interaction, robot hand-eye coordination, etc. In addition, the depth map can be further processed or analyzed as needed.

[0086] The calculation area of ​​this embodiment is small, and the accuracy is higher than that of TOF technology. Taking the resolution of the RGB binocular camera as 1080P and the preset size as 128*128 as an example, when the image is directly reconstructed in depth, the search range is determined according to the limit length of the binocular camera and the reconstruction range. Here, it is assumed that the reconstruction range is 30mm-1m and the baseline length is 100mm. The calculation amount of the depth reconstruction is 1920*1080*384. When the solution of this embodiment is adopted, the palm is deeply reconstructed, and the calculation amount is n*128*128*16, which is n / 3037 when the image is directly reconstructed in depth, where n is the number of palms, which greatly reduces the calculation amount. If this embodiment only calculates the local area m of the palm, the calculation amount is n*128*128*16*(m / 128 / 128), where n is the number of palms and m is the total number of pixels in the specified calculation area. The calculation amount is m*n / (3037*128*128) when the image is directly reconstructed. <n / 3037。

[0087] Figure 5 FIG. 1 is a schematic diagram of the structure of an alignment module in an embodiment of the present invention. Figure 5 As shown, an alignment module in an embodiment of the present invention includes:

[0088] The epipolar constraint unit is used to perform epipolar constraint processing on the first hand area and the second hand area in the y-axis direction.

[0089] Specifically, if Figure 4 As shown, the y-axis is the epipolar line of the epipolar constraint. Epipolar constraint is a key concept in stereo vision. It describes that when two cameras observe the same object, the corresponding point of a point on one image on the other image must lie on a specific straight line, which is called the epipolar line. For a horizontally arranged stereo camera system, if the two cameras are perfectly aligned horizontally and their intrinsic parameters are the same, the epipolar line is parallel to the horizontal axis (x-axis) of the image plane.

[0090] In practical applications, the epipolar line may not be completely horizontal because the camera may have a slight pitch and yaw angle. Therefore, when aligning, it is necessary to ensure that the epipolar constraints of the first hand region and the second hand region in the vertical direction (y-axis) are satisfied, that is, to ensure that the corresponding feature points in the two images have the same position in the vertical direction or follow the epipolar geometry. The epipolar constraint is usually a transformation supported by the ISP, and can also be calculated by image mapping.

[0091] The alignment unit is used to align the first hand area and the second hand area to have the same pixel size range with the center point in the x-axis direction.

[0092] Specifically, if Figure 4 As shown, the x-axis is the axis of the image perpendicular to the y-axis. After the epipolar constraint processing, the next step is to align the first hand region and the second hand region in the x-axis direction. When searching for corresponding points in the two images, the search range is limited to the same horizontal scan line, thereby greatly reducing the amount of calculation. The alignment unit ensures that the first hand region and the second hand region are aligned in the x-axis direction with their center points as the reference. This helps to reduce errors caused by the position offset of the hand in the image. During execution, first, the center points of the first hand region and the second hand region need to be determined respectively. This can be done by calculating the center of the hand region bounding box, the centroid of the hand contour, or other appropriate methods. Then, based on these center points, ranges of the same pixel size are selected in the two hand regions as the objects of subsequent processing.

[0093] Figure 6 FIG. 1 is a schematic diagram of the structure of a depth calculation module in an embodiment of the present invention. Figure 6As shown, a depth calculation module in an embodiment of the present invention includes:

[0094] A palm recognition unit is used to obtain a first palm area in the first hand area and obtain a second palm area in the second hand area.

[0095] Specifically, the palm area is further extracted from the detected hand area. Figure 7 As shown, the palm region and the finger region constitute the hand region. The palm region can usually be identified by analyzing features such as the hand contour, skin color distribution, and the geometric relationship between the fingers and the palm. For example, image segmentation techniques (such as threshold-based segmentation, region-based segmentation, or edge-based segmentation) can be used to separate the palm and finger parts. Once the palm region is extracted, these regions are marked using bounding boxes, masks, or other shapes for subsequent processing.

[0096] The first disparity unit is used to estimate the search range according to the size of the first palm area or the second palm area, and calculate the disparity of the first palm area and the second palm area.

[0097] Specifically, first, the search range in another view is estimated based on the size of the palm area, that is, the possible position range of the corresponding points of the palm area of ​​the first image is found in the second image. Then, a stereo matching algorithm (such as block matching, feature matching, etc.) is used to find the corresponding points between the first palm area and the second palm area, and their position difference in the horizontal direction, that is, disparity, is calculated. The search range is adjusted according to the size of the first palm area or the second palm area, but it is usually relatively small, such as 16, 24 pixels, etc.

[0098] The first depth unit is used to calculate the depth value of each pixel in the first palm area according to the disparity and the parameters of the binocular camera.

[0099] Specifically, the disparity value is converted into depth information using the binocular vision principle. Based on the camera's internal parameters (focal length, photosensitive element size, etc.) and external parameters (baseline distance, camera relative posture, etc.), as well as the disparity and triangulation principles, the actual depth value of each pixel relative to the camera can be calculated. In this way, we can obtain the three-dimensional position information of all pixels in the first palm area, thereby realizing the depth perception of the palm.

[0100] Figure 8 FIG. 1 is a schematic diagram of the structure of another depth calculation module in an embodiment of the present invention. Figure 8 As shown, another depth calculation module in an embodiment of the present invention includes:

[0101] The finger recognition unit is used to obtain a first finger area in the first hand area and obtain a second finger area in the second hand area.

[0102] Specifically, it is similar to extracting the palm region, but the goal is to identify and separate the finger parts in the hand image. This is usually more complicated because the fingers may have different lengths, degrees of curvature, and occlude each other. This can be achieved using methods based on shape analysis, contour tracking, or deep learning. Once the finger regions are identified, they are marked using appropriate methods (such as bounding boxes, masks, etc.) for subsequent processing.

[0103] The second disparity unit is used to determine the number of each finger in the first finger area and the second finger area, and calculate the disparity for the fingers with the same number.

[0104] Specifically, the detected fingers need to be numbered first so that the corresponding fingers can be found in the two views. The numbering can be done based on the position, size or other features of the fingers. Then, a stereo matching algorithm (such as block matching, feature matching, etc.) is used to find the corresponding points between the first finger area and the second finger area, and calculate their position difference in the horizontal direction, that is, disparity.

[0105] The second depth unit is used to calculate the depth value of each pixel in the first finger area according to the disparity and the parameters of the binocular camera.

[0106] Specifically, the disparity value is converted into depth information using the binocular vision principle. Based on the camera's internal parameters (focal length, photosensitive element size, etc.) and external parameters (baseline distance, camera relative posture, etc.), as well as the disparity and triangulation principles, the actual depth value of each pixel relative to the camera can be calculated. In this way, we can obtain the three-dimensional position information of all pixels in the first finger area, thereby realizing the depth perception of the finger.

[0107] This embodiment can calculate the depth of each finger individually, which provides more detailed three-dimensional information for gesture recognition and hand motion capture. For example, this embodiment can estimate the finger posture individually, and through detection between multiple frames, it can detect the finger movement state, meeting the needs of various application scenarios.

[0108] In some embodiments, the second disparity unit calculates the search range of each finger according to the width of each finger, and then performs matching to calculate the disparity. First, each finger needs to be identified and numbered in the first hand area and the second hand area. This is usually achieved through image processing technology, such as contour detection and fingertip recognition algorithm. Next, the search range in the second view is estimated based on the average width and direction of each finger. The search range of each finger is calculated separately, which can improve the accuracy of the search, reduce the search space in the stereo matching process, and improve the matching efficiency and accuracy. Use a stereo matching algorithm (such as block matching or feature matching) to find corresponding points between the first finger area and the second finger area. Since the width of the finger provides an additional constraint, the matching algorithm can be adjusted to take this feature into account, such as limiting the search range to the area defined by the finger width. After the match is successful, the horizontal position difference of each pair of matching points is recorded, that is, the disparity. The size of the disparity is related to the depth of the point. The larger the disparity, the closer the point is to the camera. Using the camera parameters and disparity information, the depth information of each finger can be calculated by triangulation. This step provides key three-dimensional data for subsequent gesture recognition or hand motion capture.

[0109] In some embodiments, the second disparity unit calculates the search range according to the maximum width of the finger, and then performs matching to calculate the disparity. In the first hand region, for each finger that has been identified and numbered, an image processing technique (such as edge detection, contour analysis, etc.) is used to detect the maximum width of the finger. This is usually done by finding the distance between the two farthest points in the finger contour, which correspond to the widest part of the finger. Based on the maximum width of each finger detected, a search range is calculated for the finger in the second hand region. This search range should be a rectangular or elliptical area centered on the estimated finger position and slightly larger than the maximum width of the finger. The size of the search range should be large enough to accommodate finger position offsets caused by changes in perspective, changes in hand posture, or depth differences, but should not be too large to avoid introducing unnecessary computational complexity and mismatching. Feature points (such as corner points, edge points, etc.) are extracted on each finger in the first hand region, and the best match of these feature points is found in the corresponding search range of the second hand region. This can be achieved by using feature matching algorithms (such as SIFT, SURF, ORB, etc.), which can find similar feature points between images and match them. Once a matching pair of feature points is found, the pixel position difference between them, i.e., disparity, is calculated. The size and direction of the disparity reflects the relative position of the finger in the two camera views, which can be used for subsequent 3D reconstruction or depth estimation.

[0110] Fig. 9 FIG. 1 is a schematic diagram of the structure of a detection module in an embodiment of the present invention. Fig. 9As shown, a detection module in an embodiment of the present invention includes:

[0111] A detection unit is used to detect a third hand region and a fourth hand region on the first image, and to detect a fifth hand region and a sixth hand region on the second image.

[0112] Specifically, multiple hand regions are detected by using computer vision techniques, such as edge detection, contour extraction, morphological operations, or deep learning models. The detection unit detects at least two hand regions on the first image and at least two hand regions on the second image. Each detected hand region should be correctly marked and identified so that pairing and disparity calculation can be performed correctly in subsequent processing.

[0113] A pairing unit is used to pair the third hand region, the fourth hand region, the fifth hand region and the sixth hand region according to the epipolar constraint.

[0114] Specifically, the epipolar constraint means that any point in the left image must be matched on the epipolar line of the right image, and vice versa. This constraint reduces the matching search space, thereby improving matching efficiency and accuracy. Using the epipolar geometry relationship, the area paired with the third hand area is determined in the fifth and sixth hand areas, and the area paired with the fourth hand area is determined in the fifth and sixth hand areas. This involves calculating the projection of each area on the corresponding epipolar line and finding the most likely matching point. At the same time, further verification and optimization are needed to ensure the accuracy of the pairing, such as by analyzing the size, shape and relative position of the hand area to confirm the validity of the pairing.

[0115] This embodiment processes multiple detected hand regions and pairs these regions based on epipolar constraints, which is critical for achieving accurate gesture recognition and depth perception, especially in complex scenarios that require processing multiple hands or fingers.

[0116] An embodiment of the present invention further provides an electronic device, comprising the palm-scanning binocular camera described above, and further comprising a processor and a memory, wherein executable instructions of the processor are stored.

[0117] As described above, this embodiment recognizes the hand area, scales the hand area, and obtains the depth value of the hand area by calculating the parallax, which greatly reduces the amount of calculation for search and reconstruction. There is no need to use a dedicated Asic chip or dedicated hardware, and the depth of the hand is calculated in real time to provide assistance for subsequent gesture recognition.

[0118] Fig.10 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. Fig.10The electronic device 600 according to this embodiment of the present invention is described. Fig.10 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0119] like Fig.10 As shown, the electronic device 600 is in the form of a general 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.

[0120] The storage unit stores program codes, and the program codes can be executed by the processing unit 610 .

[0121] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0122] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a grid environment.

[0123] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0124] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device 600, and / or any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more grids (e.g., a local area network (LAN), a wide area network (WAN), and / or a public grid, such as the Internet) through a 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 Fig.10Not shown, other hardware and / or software modules may be used in conjunction with 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.

[0125] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the embodiments can be referred to each other. The above description of the disclosed embodiments enables professionals and technicians in this field to implement or use the present invention. Various modifications to these embodiments will be obvious to professionals and technicians in this field, and 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 in this article, but will comply with the widest range consistent with the principles and novel features disclosed herein.

[0126] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A palm-swiping binocular camera, characterized in that: include: A first receiver, used for receiving a first signal of a palm and generating a first image; A second receiver, used for receiving a second signal of the palm and generating a second image; A processor is used to detect a first hand area on the first image, detect a second hand area on the second image, align the first hand area with the second hand area, scale the first hand area and the second hand area to a preset size at the same magnification, and then perform depth calculation on the first hand area and the second hand area.

2. A palm-swiping binocular camera according to claim 1, characterized in that: Also includes: A light source is used to illuminate the palm according to the brightness of the first image.

3. The palm-swiping binocular camera according to claim 1, characterized in that: The processor comprises: A detection module, configured to detect a first hand region on the first image and a second hand region on the second image; an alignment module, configured to align the first hand region with the second hand region; A scaling module, used for scaling the first hand area and the second hand area to a preset size at the same magnification; A depth calculation module is used to perform depth calculation on the first hand area and the second hand area.

4. The palm-swiping binocular camera according to claim 3, characterized in that: The alignment module comprises: an epipolar constraint unit, configured to perform epipolar constraint processing on the first hand region and the second hand region in the y-axis direction; The alignment unit is used to align the first hand area and the second hand area to have the same pixel size range with the center point in the x-axis direction.

5. The palm-swiping binocular camera according to claim 3, characterized in that: The depth calculation module includes: A palm recognition unit, configured to obtain a first palm region in the first hand region and obtain a second palm region in the second hand region; A first disparity unit, configured to estimate a search range according to a size of the first palm area or the second palm area, and calculate a disparity between the first palm area and the second palm area; The first depth unit is used to calculate the depth value of each pixel in the first palm area according to the disparity and the parameters of the binocular camera.

6. The palm-swiping binocular camera according to claim 3, characterized in that: The depth calculation module includes: A finger recognition unit, used to obtain a first finger region in the first hand region and a second finger region in the second hand region; A second disparity unit, used to determine the number of each finger in the first finger area and the second finger area, and calculate the disparity for fingers with the same number; The second depth unit is used to calculate the depth value of each pixel in the first finger area according to the disparity and the parameters of the binocular camera.

7. The palm-swiping binocular camera according to claim 6, characterized in that: The second disparity unit calculates the search range of each finger according to the width of each finger, and then performs matching to calculate the disparity.

8. The palm-swiping binocular camera according to claim 6, characterized in that: The second disparity unit calculates a search range according to the maximum width of the finger, and then performs matching to calculate the disparity.

9. The palm-swiping binocular camera according to claim 3, characterized in that: The detection module comprises: a detection unit, configured to detect a third hand region and a fourth hand region on the first image, and to detect a fifth hand region and a sixth hand region on the second image; A pairing unit is used to pair the third hand region, the fourth hand region, the fifth hand region and the sixth hand region according to the epipolar constraint.

10. An electronic device, characterized in that: It comprises the palm-swiping binocular camera as described in any one of claims 1 to 9.

Images