论文部分内容阅读
该系统以飞思卡尔系列单片机K60为主控芯片,CMOS摄像头0V7620作为图像传感器,将采集的手势图像送到单片机进行图像压缩、去噪滤波、动态阈值二值化、轮廓跟踪、指尖检测等处理步骤,完成手势的识别.并将识别结果通过语音模块输出,根据识别结果实现风扇的调速控制.手势轮廓的提取采用八邻域跟踪算法,指尖检测采用K_COS算法,并结合待测点与质心的距离,剔除指谷的影响.该项目不仅完成对特定手势的识别和风扇的调速控制,而且实时识别过程中脱离计算机,使用方便小巧的单片机,实现良好的人机交互,让手势控制成为了更为大众化的一种控制方法.,The system uses Freescale MCU K60 as the main control chip and CMOS camera OV7620 as the image sensor.The collected gesture images are sent to the MCU for image compression,denoising filtering,dynamic threshold binarization,contour tracking,fingertip detection and so on,which completes the recognition of the gesture.Then the recognition result is output through the voice module,and the speed regulation control of the fan is realizedaccording to the recognition result.The eight-field tracking algorithm is used to extract the gesture contours.The K_COS algorithm is applied to the fingertip detection,and the influence of the finger’svalley is eliminated by combining the distance between the measured points and the center of mass.The project,with the convenient single chip microcomputer rather than the computer,not only completes the recognition of specific gesture and the speed control of the fan,but also realizes good human-computer interaction,which makes the gesture controlbecome a more popular control methodin the process of realtime recognition.