基于改进Hough变换的指针式仪表自动识别方法研究

在对指针式仪表进行识别的过程中,Hough变换被广泛应用于检测指针的位置,但其存储空间大、计算时间长的缺点让识别的过程效率降低很多。针对传统Hough变换的缺点提出了一种改进的Hough变换,通过减小Hough变换的搜索范围来减小其存储空间和计算量,利用指针与表盘圆心之间的关系限定了Hough变换的累加量,从而大大减少了计算时间。将改进的Hough变换应用于指针式仪表的识别中,与传统的Hough变换识别速度进行了比较。实验结果表明:改进后的Hough变换可有效减少计算时间,提高识别效率。

Research on Automatic Recognition Based on Improved Hough Transform for Pointer Instrument

In the process of recognizing the pointer instrument, Hough transform is widely used to detect the position of the pointer,but disadvantages of the large storage space and the long calculating time make the recognition process inefficient. This paper studies the traditional Hough transform,proposes an improved Hough transform for its shortcomings. By reducing the searching scope of the Hough transform it reduces its storage space and computational complexity, and by the relationship between the pointer and dial center it limits the cu-mulative amount of the Hough transform,which significantly reduces the computation time. It applies the improved Hough transform to the pointer instrument of recognition, and compares with the traditional Hough transform recognition speed. Experimental results show that the improved Hough transform can effectively reduce the computation time and improve the recognition efficiency.