Method for identifying motion state of artificial limb wearer

Abstract

The invention discloses a method for identifying the motion state of an artificial limb wearer. The method comprises the following steps: 1, acquiring a signal; 2, preprocessing the initial signal based on wavelet transform; 3, extracting characteristic values of the signals based on wavelet packets; and 4, classifying and identifying the motion state based on a K-nearest neighbor algorithm. According to the invention, automatic identification of the motion state of the artificial limb wearer is realized, and the identification rate and real-time performance of motion state identification of the artificial limb wearer are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of pattern recognition, and in particular to a method for recognizing the movement state of a prosthesis wearer. Background Technology

[0002] Limb disability not only deprives patients of their ability to work but can also impair their self-care abilities, impacting their lives and those of others, and increasing the burden on families and society. With the increasing number of people with disabilities, using prostheses to assist them in their work and daily lives and help them regain mobility is a widely accepted solution. As a key component for amputees to integrate into normal life, prostheses are an important guarantee for their reintegration into society. Motion state recognition is a crucial element in intelligent prosthetic technology. By using sensors and algorithms to identify the wearer's motion state, the adaptability, control precision, and motion efficiency of prostheses can be improved, expanding the application range of intelligent prostheses and enabling wearers to control their prostheses. Therefore, motion state recognition technology for prostheses is particularly important for the development of intelligent prosthetic technology.

[0003] Traditional methods for recognizing the movement status of prostheses suffer from low recognition rates and poor real-time performance. Summary of the Invention

[0004] In view of the above situation, in order to overcome the shortcomings of traditional prosthesis wearer motion state recognition methods, such as low recognition rate and poor real-time performance, this invention proposes a prosthesis wearer motion state recognition method.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for recognizing the movement state of a prosthesis wearer includes the following steps:

[0007] (1) Signal acquisition;

[0008] (2) Preprocess the initial signal based on wavelet transform;

[0009] (3) Extracting feature values ​​from signals based on wavelet packet pairs;

[0010] (4) Classify and identify motion states based on the K nearest neighbor algorithm.

[0011] Compared with the prior art, the beneficial effects of the present invention are:

[0012] This invention uses wavelet transform to preprocess the initial signal, effectively eliminating noise. It utilizes wavelet packets to analyze the signal and extract feature values, adaptively selecting the frequency band based on the characteristics of the analyzed signal, thus improving signal processing capabilities. The K-nearest neighbor algorithm is used for motion state recognition, achieving high recognition rate and strong real-time performance. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0014] The present invention will now be described in conjunction with the accompanying drawings and specific embodiments.

[0015] This invention proposes a method for recognizing the movement state of a prosthesis wearer, which specifically includes the following steps.

[0016] Step 1: Signal Acquisition.

[0017] Initial signals from the triaxial MEMS accelerometer and triaxial MEMS gyroscope placed inside the prosthesis receiving cavity were acquired.

[0018] Step 2: Preprocess the initial signal based on wavelet transform.

[0019] The initial signal is decomposed using wavelet transform. The noise standard deviation is estimated from the high-frequency coefficients of the first layer. A global threshold is used for noise reduction, and a soft threshold is used to perform noise reduction on the saved low-frequency signal.

[0020] Step 3: Extract feature values ​​from the signal based on wavelet packets.

[0021] The signal is decomposed using db1 wavelet packets to obtain decomposed signals of different frequency bands of the acceleration and angular velocity signals, and the power of the signals is extracted. The standard deviation of the reconstructed low-frequency signal is selected as the signal feature value to generate a feature vector.

[0022] Step 4: Classify and identify motion states based on the K-nearest neighbor algorithm.

[0023] Calculate the Euclidean distance between the training samples and the feature vector to be identified, take the K samples with the smallest distance, and determine the category to which these samples belong. This category is the identified motion state.

[0024] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for recognizing the movement state of a prosthesis wearer, characterized in that: Includes the following steps: (1) Signal acquisition; (2) Preprocess the initial signal based on wavelet transform; (3) Extracting feature values ​​from signals based on wavelet packet pairs; (4) Classify and identify motion states based on the K nearest neighbor algorithm.

Images