中文说明：
首先利用形态学开运算将图像中小于结构元素的亮点去掉，利用源图像与处理后的图像作差，对差图像进行阈值分割，得到二值图像，将二值图像中的一些干扰区域去掉，提取二值图像中的圆形区域，并在源图像上对小目标进行标记，实现对圆形亮目标的检测。
English Description：
Firstly, the bright spots smaller than the structural elements in the image are removed by morphological operation, and the difference between the source image and the processed image is used for threshold segmentation to get the binary image. Some interference areas in the binary image are removed, and the circular areas in the binary image are extracted, and the small targets are marked on the source image to realize the detection of the circular bright targets.