基于非亚采样轮廓波的舰船尾迹检测

介绍了海洋表面合成孔径雷达(SAR)图像的舰船尾迹检测算法.该算法首先对图像进行对数变换,将乘性噪声转化为加性噪声,然后对图像进行非亚采样contourlet变换,根据变换后的像素均值对变换系数进行调整,以便去除噪声和提取尾迹轮廓.接着对经非亚采样contoulet调整后的图像进行局部Radon变换,即在对像素积分过程中沿着线性特征被分割后的若干短线段进行积分,而非对整幅图像进行积分.最后通过采用形态学膨胀和腐蚀的方法将临近的分离短线段连接起来便于后续的处理.采用该算法对实际SAR图像进行舰船尾迹检测,结果表明:该算法在背景噪声很强时仍能检测出略微有弯曲的尾迹信号.

Ship wake detection based on nonsubsampled contourlet

The development of an algorithm to automatically detect ship wakes in digital synthetic aperture radar(SAR)images of the ocean surface is described．Firstly，this algorithm performs a logarithmic transform to convert the multiplicative speckle noise into additive noise，and then performed the nonsubsampled contourlet transform．The coefficients are classified to two types according to the mean value．and then modified to get rid of the noise and check the contours of the ship wakes．The second key element of this technique is a localization of the Radon transform，whereby the intensity integration is performed over short line segments rather than across the entire image．Lastly，the algorithm applies the dilation and erosion mathematical morphology of binary images to the image obtained in order to connect the adjacent line segments．This algorithm is tested by actual SAR images of ship wakes．The resultsdemonstrate that this algorithm's robustness in the presence of noise．as well as its ability to detect and localize linear features that are somewhat not so straight．作者简介：孙继银(1952-),男,山东单县人,教授,博士生导师,研究方向为图像处理、导弹制导.Email:grass2009@163.com