Hand motion trail recognition method, system and equipment and medium
By improving the YOLOv5 gesture detection model and optimizing inertial navigation data, the robustness and efficiency issues of hand motion trajectory acquisition in existing technologies have been resolved, achieving fast and accurate trajectory recognition while reducing system complexity and cost.
Patent Information
- Application Number
- CN202511302836.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-12
- Publication Date
- 2026-01-27
AI Technical Summary
Existing hand motion trajectory acquisition technologies are not robust enough in complex backgrounds, have high equipment costs, complex system deployment, and are difficult to achieve non-contact acquisition. Furthermore, multi-camera calibration is cumbersome and has high environmental requirements, resulting in low recognition efficiency and poor operability.
An improved YOLOv5 gesture detection model is adopted, which combines the SimAM attention mechanism and the C3Rep module to quickly locate the hand region. The hand movement trajectory is accurately identified and fitted by optimizing the pre-trained hand trajectory recognition model and inertial navigation data.
It improves the speed and accuracy of hand motion trajectory recognition, provides reliable data support, reduces computational load and equipment costs, and enhances the robustness and simplicity of the system.
Smart Images
Figure CN121415459A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of computer vision and human-computer interaction technology, and in particular to a method, system, device and medium for recognizing hand movement trajectories. Background Technology
[0002] Hand motion tracing plays a crucial role in applications such as rehabilitation therapy, telemedicine, and human-computer interaction. Existing hand motion trajectory acquisition technologies are mainly classified into two categories: First, sensor-based data acquisition methods. These methods typically acquire hand motion data directly through sensors or use optical motion capture systems that utilize multiple infrared cameras to simultaneously capture reflective points placed on the hand, thereby reconstructing the three-dimensional motion trajectory. However, these methods are usually expensive, have complex system deployments, are highly sensitive to the external environment, and are prone to data loss or errors due to occlusion of marker points, making them insufficiently robust in complex backgrounds. They also have a heavy data processing burden and are difficult to implement for non-contact acquisition, limiting their use in rehabilitation scenarios. Second, machine vision-based data acquisition methods. These methods integrate image information from multiple perspectives or cameras through multi-view visual fusion to output the motion trajectory, or utilize deep learning networks to denoise, fill in gaps, and align the raw data to obtain the hand motion trajectory. However, these methods often suffer from problems such as cumbersome multi-camera calibration, high environmental requirements, blind spots, and low recognition efficiency. In addition, these systems often require regular calibration and maintenance, which is technically challenging and makes them less feasible in large-scale customized hand motion trajectory acquisition scenarios. Summary of the Invention
[0003] Therefore, it is necessary to provide a method, system, device, and medium for hand motion trajectory recognition to address the aforementioned technical problems. This method can quickly locate the hand area, accurately identify and fit the hand motion trajectory, thereby improving the speed and accuracy of trajectory recognition and providing reliable data support for related technologies.
[0004] Firstly, a method for recognizing hand movement trajectories is provided, including: Obtain hand-captured image data; The hand image data is input into the improved YOLOv5 gesture detection model to obtain the hand region. The improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network. The C3 module in the backbone network and the neck network is replaced with the C3Rep module. The SimAM attention mechanism is introduced into the backbone network. Based on the described hand area, the hand movement trajectory is obtained.
[0005] Furthermore, the hand-captured image data is a regular RGB image or video stream.
[0006] Further, the C3Rep module comprises: a first convolutional layer, a reparameterized VGG network, an addition layer, a connection layer and a second convolutional layer connected in sequence, wherein the first convolutional layer is directly connected to the addition layer; a third convolutional layer connected between the first convolutional layer and the connection layer.
[0007] Further, the hand motion trajectory is obtained according to the hand region, comprising: inputting the hand recognition region into a pre-trained hand trajectory recognition model to obtain the hand motion trajectory.
[0008] Further, the hand motion trajectory is obtained by inputting the hand recognition region into a pre-trained hand trajectory recognition model, comprising: inputting the hand recognition region into a pre-trained hand trajectory recognition model to detect a plurality of hand key points; analyzing the plurality of hand key points into a time sequence; interpolating between the corresponding hand key points based on the time sequence; according to the interpolation result, applying a least square fitting method for trajectory fitting to obtain a smooth hand motion trajectory.
[0009] Further, the interpolation between the corresponding hand key points based on the time sequence comprises: applying a polynomial interpolation method to interpolate between the corresponding hand key points; or applying a spline interpolation or RANSAC algorithm to interpolate between the corresponding hand key points.
[0010] Further, it further comprises: obtaining inertial navigation data and optimizing the hand motion trajectory according to the inertial navigation data.
[0011] an acquisition module for obtaining hand shooting image data; a detection module for inputting the hand shooting image data into an improved YOLOv5 gesture detection model to obtain a hand region, wherein the improved YOLOv5 gesture detection model comprises a backbone network, a neck network and a head network, a C3 module in the backbone network and the neck network is replaced by a C3Rep module, and a SimAM attention mechanism is introduced in the backbone network; an identification module for obtaining a hand motion trajectory according to the hand region.
[0012] In a third aspect, a computer device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the steps of the hand motion trajectory recognition method of the first aspect and any possible implementation manner of the first aspect are implemented.
[0013] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and when the processor executes the program, the steps of the hand motion trajectory recognition method of the first aspect and any possible implementation manner of the first aspect are implemented.
[0014] By adopting the embodiment of the present application, first, hand shooting image data is obtained; then the hand shooting image data is input into the improved YOLOv5 gesture detection model to obtain a hand region, wherein the improved YOLOv5 gesture detection model includes a backbone network, a neck network and a head network, the C3 module in the backbone network and the neck network is replaced with a C3Rep module, and the SimAM attention mechanism is introduced into the backbone network; finally, the hand motion trajectory is obtained according to the hand region. Therefore, the hand region can be quickly located, the hand motion trajectory can be accurately identified and fitted, so as to improve the speed and accuracy of trajectory recognition, and to further provide reliable data support for related technologies. BRIEF DESCRIPTION OF DRAWINGS
[0015] Other characteristics, objects and advantages of the present application will become more apparent from the following detailed description of the non-restrictive embodiments made with reference to the accompanying drawings: Figure 1 A flowchart of the hand motion trajectory recognition method provided by the embodiment of the present application is provided; Figure 2 An algorithm flowchart of the hand motion trajectory recognition method provided by the embodiment of the present application is provided; Figure 3 A network structure diagram of the improved YOLOv5 provided by the embodiment of the present application is provided; Figure 4 A C3Rep network structure diagram provided by the embodiment of the present application is provided; Figure 5 A SimAM attention mechanism working principle diagram provided by the embodiment of the present application is provided; Figure 6 A hand key point position schematic diagram provided by the embodiment of the present application is provided; Figure 7 A trajectory fitting flowchart in the hand motion trajectory recognition method provided by the embodiment of the present application is provided; Figure 8 A structure block diagram of the hand motion trajectory recognition system provided by the embodiment of the present application is provided; Figure 9 A structure block diagram of the computer device provided by the embodiment of the present application is provided. Detailed Implementation
[0016] The present application will now be described in further detail with reference to the embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the application. Furthermore, it should be noted that, for ease of description, only the parts relevant to the application are shown in the accompanying drawings.
[0017] It should be noted that, unless otherwise specified, the embodiments and features of the embodiments in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.
[0018] The following describes in detail, with reference to the accompanying drawings, a method, system, device, and medium for recognizing hand movement trajectories according to embodiments of this application.
[0019] Figure 1 This is a flowchart of a hand movement trajectory recognition method according to an embodiment of this application. Figure 1 As shown, the hand movement trajectory recognition method according to an embodiment of this application includes the following steps: S101: Acquire hand-captured image data.
[0020] In one embodiment of this application, the hand-captured image data is a standard RGB image or video stream. First, the input image or video stream undergoes data preprocessing, specifically including scaling and pixel normalization.
[0021] Scaling refers to adjusting the input RGB image or video stream to a uniform size so that the model can process it in batches. For example, in a specific example, the input data size is uniformly adjusted to 640×640. Pixel normalization refers to mapping the original pixel values (0-255) of the input data to the range [0,1], thereby ensuring the stability of the model's input distribution and avoiding gradient explosion or slow convergence.
[0022] S102: Input the hand image data into the improved YOLOv5 gesture detection model to obtain the hand region. The improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network. The C3 module in the backbone network and the neck network is replaced with the C3Rep module. The backbone network introduces the SimAM attention mechanism.
[0023] Improve the YOLOv5 network structure, such as Figure 3As shown, the improved YOLOv5 network is composed of a backbone network, a neck network and a head network, wherein the backbone network is responsible for feature extraction of the preprocessed hand shooting image data, so as to convert the input image data into feature maps of different scales; the neck network is responsible for feature fusion, so that the feature map information of different scales interacts to enhance the target detection capability; and the head network is responsible for target detection of the fused feature map and outputs the result.
[0024] In an embodiment of the present application, the C3Rep module comprises a first convolutional layer, a reparameterized VGG network, an addition layer, a connection layer and a second convolutional layer connected in sequence, wherein the first convolutional layer is directly connected with the addition layer; and a third convolutional layer connected between the first convolutional layer and the connection layer.
[0025] The network structure of the C3Rep module is as shown in Figure 4 As shown, the C3Rep module is improved by combining the C3 module with the reparameterized convolution, that is, the original Bottleneck structure is replaced by the reparameterized VGG structure. Specifically, in the training stage, a multi-branch convolutional structure is adopted, such as 3x3 convolution, 1x1 convolution and Identity branch, to enhance the feature learning capability; in the inference stage, the branches are converted into a single-branch convolutional structure through reparameterization, so as to improve the inference speed and reduce the calculation amount. Through this method, the expression capability of feature extraction can be enhanced while maintaining the lightweight of the model.
[0026] On the other hand, the improved YOLOv5 network introduced by the present application introduces a SimAM attention mechanism. The working principle of SimAM is as shown in Figure 5 As shown, the SimAM attention mechanism obtains a three-dimensional weight of the original feature, then extends the weight to the spatial dimension, and fuses it with the original feature, so as to highlight the key information points without increasing the additional parameter amount.
[0027] Through the improved YOLOv5 network described above, the original RGB image or video stream data can be positioned for hand bounding box, and at the same time it is ensured that only a single hand region is included in the data, and if multiple hands are detected, the processing will be stopped. The specific hand detection process is as shown in Figure 2 As tested, compared with the original YOLOv5 network, the improved YOLOv5 network proposed in the present application reduces the parameter amount by 34.5%, reduces the calculation amount by 34.9%, reduces the inference time by 0.7ms, and at the same time, the gesture detection accuracy reaches 98.5%.
[0028] S103: obtaining a hand motion trajectory according to the hand region.
[0029] In one embodiment of this application, obtaining the hand movement trajectory based on the hand region includes: inputting the hand recognition region into a pre-trained hand trajectory recognition model to obtain the hand movement trajectory.
[0030] In one embodiment of this application, combined with Figure 2 As shown, the step of inputting the hand recognition region into a pre-trained hand trajectory recognition model to obtain the hand movement trajectory includes: inputting the hand recognition region into the pre-trained hand trajectory recognition model to detect multiple hand key points; parsing the multiple hand key points into a time series; interpolating between the corresponding hand key points based on the time series; and applying the least squares fitting method to fit the trajectory according to the interpolation result to obtain a smooth hand movement trajectory.
[0031] The process of fitting the trajectory is as follows Figure 7 As shown, in this application, the MediaPipe Hands model is used for hand trajectory recognition. The hand recognition area obtained in the above process is input into the recognition model, and key point detection is used to obtain the following... Figure 6 The diagram shows several key points. By analyzing the temporal changes in the positions of these key points, it can be analyzed as a time series.
[0032] In one embodiment of this application, the interpolation between corresponding hand key points based on the time series includes: applying polynomial interpolation to interpolate between corresponding hand key points; or applying spline interpolation or the RANSAC algorithm to interpolate between corresponding hand key points.
[0033] Specifically, during the recognition process, there may be issues such as missing or unevenly spaced key points on the hand. To form a continuous and smooth trajectory curve, interpolation is required. Assume the time series contains... n +1 data point For this, construct polynomials of degree no more than n, as shown in Formula 1 and Formula 2: (1) (2) in, P n (x) It is an interpolation polynomial; Li(x) It is a Lagrange function.
[0034] It should be noted that in other examples, spline interpolation or the RANSAC algorithm can also be used to perform interpolation between the corresponding hand key points.
[0035] Based on interpolation, since the hand movement trajectory data may contain noise and jitter, the least squares fitting method is applied to further fit the trajectory data in order to obtain a stable and smooth trajectory curve. The specific process is shown in Equation 3: (3) in, m The number of sampling points; Q k (x i ) for k Polynomial of degree 1; E This represents the fitting error.
[0036] By solving the polynomial coefficients through the normal equation system to minimize the fitting error, a smooth fitting curve that most closely approximates the actual trajectory can be obtained.
[0037] By fusing polynomial interpolation with least squares fitting, the continuity and robustness of the data can be improved. In practice, the time series is segmented according to motion characteristics, and the above interpolation and fitting operations are performed on each segment to ensure the continuity of the curve.
[0038] In one embodiment of this application, the method further includes: obtaining inertial navigation data and optimizing the hand movement trajectory based on the inertial navigation data.
[0039] Specifically, depth perception can be aided by sensors to obtain inertial navigation data. By using the inertial navigation data to optimize the hand motion trajectory, the accuracy of 3D trajectory detection and recognition can be improved.
[0040] According to the hand movement trajectory recognition method of this application, firstly, hand image data is obtained; then, the hand image data is input into an improved YOLOv5 gesture detection model to obtain the hand region. The improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network. The C3 module in the backbone network and the neck network is replaced with a C3Rep module, and a SimAM attention mechanism is introduced into the backbone network. Finally, the hand movement trajectory is obtained based on the hand region. Therefore, the hand region can be quickly located, and the hand movement trajectory can be accurately identified and fitted, thereby improving the speed and accuracy of trajectory recognition and providing reliable data support for related technologies.
[0041] Figure 8 This is a structural block diagram of a hand movement trajectory recognition system according to an embodiment of this application. Figure 8 As shown, a hand movement trajectory recognition system according to an embodiment of this application includes: an acquisition module 810, a detection module 820, and a recognition module 830, wherein: The acquisition module 810 is used to acquire hand-captured image data; The detection module 820 is configured to input the hand photograph data into an improved YOLOv5 gesture detection model to obtain a hand region, wherein the improved YOLOv5 gesture detection model comprises a backbone network, a neck network and a head network, a C3 module in the backbone network and the neck network is replaced by a C3Rep module, and a SimAM attention mechanism is introduced into the backbone network. The recognition module 830 is configured to obtain a hand motion trajectory according to the hand region.
[0042] According to the hand motion trajectory recognition system, the hand photograph image data is obtained first, then the hand photograph data is input into the improved YOLOv5 gesture detection model to obtain the hand region, wherein the improved YOLOv5 gesture detection model comprises the backbone network, the neck network and the head network, the C3 module in the backbone network and the neck network is replaced by the C3Rep module, and the SimAM attention mechanism is introduced into the backbone network, and finally the hand motion trajectory is obtained according to the hand region. Therefore, the hand region can be quickly positioned, the hand motion trajectory can be accurately recognized and fitted, the speed and the accuracy of the trajectory recognition are improved, and reliable data support is provided for related technologies.
[0043] The specific limitation of the hand motion trajectory recognition system can be referred to the limitation of the hand motion trajectory recognition method in the foregoing, and details are not described herein. The modules of the hand motion trajectory recognition system can be realized by software, hardware and combinations thereof. The modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to be called and executed by the processor.
[0044] Reference is made below to Figure 9 , Figure 9 A structure schematic diagram of a computer device suitable for implementing the embodiments of the present application is shown.
[0045] As shown in Figure 9 , the computer system 1000 comprises a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage part 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for operation instructions of the system are also stored. The CPU 1001, the ROM 1002 and the RAM 1003 are connected to each other through a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.
[0046] The following components are connected to the I / O interface 1005: an input part 1006 including a keyboard, a mouse, etc.; an output part 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage part 1008 including a hard disk, etc.; and a communication part 1009 including a network interface card such as a LAN card, a modem, etc. The communication part 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as necessary. A removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 1010 as necessary, so that a computer program read out therefrom is installed in the storage part 1008 as necessary.
[0047] In particular, according to embodiments of the present application, the above-mentioned processes with reference to the flowcharts Figure 1 The described processes can be implemented as computer-readable storage media. For example, embodiments of the present application include a computer-readable storage medium including a computer program containing program code for executing the methods shown in the flowcharts, for example, for performing: obtaining hand photograph image data; inputting the hand photograph image data into an improved YOLOv5 gesture detection model to obtain a hand region, wherein the improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network, C3 modules in the backbone network and the neck network are replaced with C3Rep modules, and a SimAM attention mechanism is introduced into the backbone network; and obtaining a hand motion trajectory according to the hand region.
[0048] In particular, according to embodiments of the present application, the above-mentioned processes with reference to the flowcharts Figure 1 The described processes can be implemented as computer software programs. For example, embodiments of the present application include a computer program product including a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the flowcharts, for example, for performing: obtaining hand photograph image data; inputting the hand photograph image data into an improved YOLOv5 gesture detection model to obtain a hand region, wherein the improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network, C3 modules in the backbone network and the neck network are replaced with C3Rep modules, and a SimAM attention mechanism is introduced into the backbone network; and obtaining a hand motion trajectory according to the hand region.
[0049] In such embodiments, the computer program comprises program code for executing the methods illustrated in the flowcharts. In such embodiments, the computer program can be downloaded and installed from the network via the communication section 1009, and / or installed from the detachable medium 1011. When the computer program is executed by the central processing unit (CPU) 1001, the above-described functions defined in the system of the present application are executed.
[0050] It should be noted that the computer readable medium shown in the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium that can send, propagate or transmit the program for use by or in connection with an instruction execution system, device or apparatus. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.
[0051] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flow diagrams and / or block diagrams.
[0052] The units or modules described in the embodiments of the present application can be implemented by software, or by hardware. The units or modules described can also be implemented by a processor. In some cases, the names of the units or modules do not constitute a limitation on the units or modules themselves.
[0053] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but it should be considered that any combination of the technical features is within the scope of the present application, as long as the combination does not result in contradictions.
[0054] The above embodiments only express several implementation manners of the present application, and the description is specific and detailed, but it should not be considered as a limitation on the patent scope of the present application. It should be noted that, for those skilled in the art, some modifications and improvements can be made without departing from the concept of the present application, and these modifications and improvements are also within the protection scope of the present application. Therefore, the patent protection scope of the present application should be subject to the appended claims.
Claims
1. A method for recognizing hand movement trajectories, characterized in that, include: Obtain hand-captured image data; The hand image data is input into the improved YOLOv5 gesture detection model to obtain the hand region. The improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network. The C3 module in the backbone network and the neck network is replaced with the C3Rep module. The SimAM attention mechanism is introduced into the backbone network. Based on the described hand area, the hand movement trajectory is obtained.
2. The hand movement trajectory recognition method according to claim 1, characterized in that, The hand image data is a regular RGB image or video stream.
3. The hand movement trajectory recognition method according to claim 1 or 2, characterized in that, The C3Rep module includes: The network consists of a first convolutional layer, a reparameterized VGG network, an additive layer, a connecting layer, and a second convolutional layer, which are connected in sequence. The first convolutional layer is also directly connected to the additive layer. A third convolutional layer is connected between the first convolutional layer and the connecting layer.
4. The hand movement trajectory recognition method according to claim 1, characterized in that, The step of obtaining the hand movement trajectory based on the hand region includes: The hand recognition region is input into a pre-trained hand trajectory recognition model to obtain the hand movement trajectory.
5. The hand movement trajectory recognition method according to claim 4, characterized in that, The step of inputting the hand recognition region into a pre-trained hand trajectory recognition model to obtain the hand movement trajectory includes: The hand recognition area is input into a pre-trained hand trajectory recognition model, and multiple hand key points are detected; The multiple key hand points were analyzed into a time series; Based on the time series, interpolation is performed between the corresponding key hand points; Based on the interpolation results, the least squares fitting method is applied to fit the trajectory in order to obtain a smooth hand movement trajectory.
6. The hand movement trajectory recognition method according to claim 5, characterized in that, The interpolation between corresponding hand key points based on the time series includes: Polynomial interpolation is applied to interpolate between the corresponding key points of the hand; or... Interpolation between the corresponding hand key points is performed using spline interpolation or the RANSAC algorithm.
7. The hand movement trajectory recognition method according to claim 1, characterized in that, Also includes: Obtain inertial navigation data and optimize the hand movement trajectory based on the inertial navigation data.
8. A hand movement trajectory recognition system, characterized in that, include: The acquisition module is used to obtain hand-captured image data; The detection module is used to input the hand image data into the improved YOLOv5 gesture detection model to obtain the hand region. The improved YOLOv5 gesture detection model includes a backbone network, a neck network, and a head network. The C3 module in the backbone network and the neck network is replaced with the C3Rep module. The SimAM attention mechanism is introduced into the backbone network. The recognition module is used to obtain the hand movement trajectory based on the hand area.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the hand movement trajectory recognition method according to any one of claims 1-7.
10. A computationally readable storage medium, comprising a memory and a computer program stored on the memory and executable on a processor, characterized in that, When the program is executed by the processor, it implements the hand movement trajectory recognition method according to any one of claims 1-7.