【摘要】 针对空中远距离红外小目标检测的实际问题,提出了一种基于小波变换的检测算法。该方法首先对小目标图像进行小波分解并在考虑高频系数能量的基础上对噪声和背景边缘系数进行抑制,然后将遗留下来的高频系数通过线性映射变换成灰度图像。其次对3个方向的高频图像按照一维最大熵法进行二值化处理并通过形态开算子进一步滤除噪声,随后将高频图像两两相与关联生成单帧检测结果,并进一步利用帧间目标位置的相关性完成小目标检测过程。最后,在原图像中以检测结果图像的质心为中心生成跟踪窗口。试验结果表明,相对于通常的小目标检测算法,提出的算法在背景抑制、检测准确度以及速度方面都具有一定的优势。更多还原

【Abstract】 In order to solve the practical problem of the infrared small target’s detecting in the sky,a detecting algorithm based on the wavelet transform was proposed.Firstly,the infrared image which included small target was decomposed by the wavelet transform.In order to suppress background and noise,the wavelet coefficients in high frequence were processed by the threshold method which took coefficient energy into account.Then the left coefficients were mapped to gray image by the linear transform.Secondly,the three mapped images in high frequency were binarized by the method of one-dimensional maximum entropy and the morphology open operator was used to filter the noises further.Subsequently,the result image of single frame was acquired by taking the correlation between the three images in high frequency into account.The final detecting result was obtained by considering the position correlation of the target between frames.In the end,the tracking window whose center was the centroid of the result image was generated in the original image.Experimental results show that the method given by this paper,compared with some traditional method,has certain advantages in background suppressing,detecting accuracy and speed.更多还原