基于改进YOLO-v3的眼机交互模型及实现

针对嵌入式眼-机交互技术中所采用的传统眼行为识别方法准确率低、速度慢等问题,并结合所研制眼机交互系统硬件特点及应用场景,提出一种基于改进YOLO-v3的眼机交互模型.该模型通过去除13×13特征分辨率的检测模块、增加浅层网络的层数以及采用K-means聚类算法选取初始先验框,提高了网络像素特征提取细粒度并加快了检测速度,进而结合人眼特征参数提取方法和眼行为识别算法,构建出了眼机交互模型并进行实验.实验结果表明,该模型对不同眼行为的识别率达91.30％,改进的YOLO-v3网络的平均检测准确率(mean average precision,mAP)为99.9％,识别速度达22.8 FPS,相比原YOLO-v3方法检测时间缩短了11.4％.

Research and implementation of eye computer interaction model based on improved YOLO-v3

In view of the low accuracy and slow speed of the traditional eye behavior recognition methods used in the embedded Eye computer interaction technology, and combined with the hardware characteristics and application scenarios of the developed Eye computer interaction system, this paper proposes an eye computer interaction model based on the improved YOLO-v3.By removing the 13×13 feature resolution detection module, increasing the number of layers of shallow network and using k-means clustering algorithm to select the initial candidate box, the model improves the fine-grained feature extraction of network pixels and speeds up the detection speed, and then combines the extraction method of human eye feature parameters and eye behavior recognition algorithm to build an eye machine interaction model and carry out experiments. The experimental results show that the recognition rate of the model is 91.30%, the mAP of the improved YOLO-v3 network is 99.9%, and the recognition speed is 22.8fps. Compared with the original YOLO-v3 method, the detection time is shortened by 11.4%.