Infant abnormal behavior detection method based on meanshift algorithm and SVM

Abstract

The invention discloses an infant abnormal behavior detection method based on a meanshift algorithm and an SVM. The method comprises the following steps: preprocessing an acquired baby video; performing target motion trail tracking on four limbs and the whole body of the baby in the video by using a meanshift algorithm; storing the obtained motion trail information; then, wavelet transform is usedfor extracting motion trail information; establishing a sample set for the extracted wavelet approximate waveform; the set SVM support vector machine is used for training; solving a power spectrum ofthe motion trail information by using wavelets; and establishing a sample set by using the obtained features, training the sample set by using the set SVM support vector machine, testing the two trained models, and setting different weight parameters by using a data weighted fusion algorithm according to different accuracies of the two models so as to carry out weighted judgment, thereby obtaining an optimal training result.

Description

technical field [0001] The invention belongs to the technical field of video image processing, and in particular relates to a method for detecting abnormal behavior of infants based on meanshift algorithm and SVM. Background technique [0002] The abnormal behavior of the baby mainly refers to the sequence of the whole body movement is small and monotonous within one month after birth, and lacks movement complexity and fluency; within two to five months after birth, there is no variable acceleration throughout the whole body. Small range of medium-speed movement in all directions, lack of other age-appropriate movement forms, such as midline movement of limbs, hand-knee touch, visual search, fingers picking clothes, etc., and poor overall movement fluency. Clinically, the persistence of abnormal behavior in infants often indicates adverse neurodevelopmental outcomes, and the possibility of developing cerebral palsy and mental retardation is extremely high. Clinically, the d...

AI Technical Summary

Problems solved by technology

The first method is to use a specific video recording method for the baby, use the whole body movement quality evaluation criteria for the video results, and judge whether the baby's behavior is normal by the person who has obtained the Prechtl's Method evaluation qualification certificate issued by the European GM Trust. This method mainly relies on clinical observation , there is a certain subjectivity

The second method is to wear a sensor device for the baby to observe the parameters, but this wearable method itself will cause some interference to the baby's movement, resulting in inaccurate prediction results

The third method is to use the computer to extract the baby's movement characteristics for pattern recognition analysis. This method will not interfere with the baby's movement and is objective, but in the process of extracting movement characteristics and recognition, often only a limited number of body parts are used. There is no analysis of the overall movement of the whole body, so it has a certain specificity

[0004] Due to the defects of the above algorithm, it is difficult to achieve ideal results in practical applications, so it is necessary to improve

Images