In the process of mobile phone shell assembly, adverse was easy to produced in its surface, such as stains, these bad flaws tend to be around camera, distance sensor, the horn, and logo, the appearance of these goals directly affects the quality of mobile phones, so for these target localization research is particular-ly important. And the current image target localization algorithm is not stable, when the target is very small;rotating which make characteristics not obvious, make poor quality of positioning. To solve this, this paper proposes a round and rectangular target positioning mechanism of mobile phone surface based on OpenCV. First of all, based on canny edge detection and hough transform to realize circular object detection; Then based on threshold segmentation and morphological processing of rectangular target area, roughly rectangular target precise coordinates is obtained by contour matching again, contour features of area, perimeter, aspect ratio, width is poor. Finally, human-computer interaction mechanism is introduced, then measuring the dis-tance between the target in the image by mouse, enhance the user experience. Finally tested in this paper, the mechanism of performance, the results show that, compared with ordinary image target localization algo-rithm in the image of the case, the rotation mechanism in this paper has better positioning effect, pinpoint the outline of the image target.%在手机外壳组装过程中，其表面易产生不良瑕疵，这些瑕疵存在于摄像头、距离感应器以及喇叭、logo目标附近，直接影响手机质量，因此对于这些目标的检测定位研究显得尤为重要。而现有的图像目标定位机制不稳定，当目标物很小，或发生旋转时，其定位质量不理想，存在较大的定位误差。故文章设计了OpenCV耦合人机交互的手机表面圆目标与矩形目标定位机制。首先基于canny边缘检测算子与霍夫变换，实现手机圆目标检测；再利用阈值分割和形态学开运算对其进行处理，获取矩形目标的基本轮廓，通过轮廓匹配，计算该目标的精确坐标；最后引入人机交互机制，即可通过鼠标测量图像中目标间距离，增强用户体验。实验结果显示：与当前图像目标定位机制相比，在图像目标存在旋转情况中，本文机制的定位精度更高，显著降低了定位偏差，可精确定位图像目标；且目标检测正确率更高。
展开▼
