Method and system for detecting gestures

Abstract

The invention provides a method for detecting gestures. The method includes steps of detecting a predefined motion mode to determine a region-of-interest to be detected; detecting in the region-of-interest according to a scheme of a multi-scale sliding window, extracting local mean features from skin color membership images based on window images at first, classifying the features through a pre-trained Ada-Boost classifier, further extracting feature points on the basis of a gray-scale image of the window images processed via the Ada-Boost classifier, and then classifying the point-pair features through a random forest classifier; and clustering different types of target gesture windows and outputting accurate positions and shapes of the gestures. The invention further provides a system for detecting the gestures. The method and the system for detecting the gestures are simple, rapid and stable in implementing, and have the advantages of real-time performance, interference resistance, high identification precision and the like.

Description

technical field [0001] The invention relates to the field of target detection and recognition based on vision and image and human-computer interaction, in particular to a gesture detection method and system suitable for a TV embedded platform. Background technique [0002] Somatosensory control technology based on gesture recognition has become an important means of human-computer interaction. It collects the user's action pictures through ordinary cameras, detects and locates the hand features in the image through the pattern recognition algorithm, and recognizes the shape of the hand, converts this recognition information into an operation signal, and feeds it back to terminals such as smart TVs , and trigger corresponding operation commands, such as switching TV programs, adjusting volume, simple game interaction, etc. Gesture recognition technology is based on the camera equipped on the smart terminal, and the above operations can be completed by installing the correspo...

AI Technical Summary

Problems solved by technology

However, 3D scanning equipment is large in size, high in hardware cost, and requires higher computing power, so it is difficult to integrate and apply it to popular smart terminals such as TVs and mobile phones;

[0006] (3) Technology based on ordinary camera 2D image recognition: Since this technology is realized based on ordinary camera, it is also the technology with the most potential for large-scale application, but the disadvantages of this technology are also very obvious: a) For purely based on skin color For gesture detection technology based on features, environmental lighting can easily change the color features of the hand, making detection difficult; b) For gesture detection based on shape features, similar target objects in complex backgrounds can easily cause errors c) For the gesture detection algorithm that integrates multiple features, although its detection accuracy has been improved, it still cannot fundamentally solve the problem of the influence of lighting and background. More importantly, with the complexity of the algorithm It is obviously difficult to meet the needs of existing smart TV and other terminals

[0007] In summary, due to the complexity of the current gesture recognition algorithm and the large amount of video processing required, it is difficult to run smoothly on the existing embedded platforms of smart terminals such as smart TVs.

Images