基于差异梯度直方图与显著性映射的红外弱小目标检测算法

Infrared dim small target detection algorithm based on difference histogram of oriented gradients and saliency map

为了提高复杂环境下的红外弱小目标的探测能力,基于人类视觉注意机制,设计了差异直方图耦合显著性映射的弱小目标检测算法。考虑真实目标与其邻域之间的强度差异,通过采用DFT变换来计算红外图像的幅度与相位频谱,以计算二者之间的频谱残差,并联合高斯滤波方法,输出显著性映射,从而有效突出显著区域。分析目标与背景区域的梯度幅度和梯度方向之间的差异,计算红外图像的差异直方图,以充分抑制背景杂波和噪声。联合显著性映射与差异直方图,形成图像对应的融合特征映射。最后,引入自适应阈值分割方法,从融合特征映射中准确定位真实目标。多组测试数据表明,较已有的红外目标检测技术而言,所提算法能够更好地定位出弱小目标,呈现出理想的ROC曲线。

In order to improve the detection ability of infrared dim small target in complex environment,by introducing the human visual attention mechanism,a weak small target detection algorithm based on difference histogram of oriented gradients coupling saliency mapping is designed.Considering the intensity difference between the real object and its neighborhood,the DFT was used to calculate the amplitude and phase spectrum of the infrared image for calculating the spectrum residual between them,and the saliency mapping was outputed by combinning with Gaussian filtering method to effectively highlight the salient region.The difference between the gradient amplitude and the gradient direction of the target and the background area was analyzed,and the difference histogram of oriented gradients of the infrared image was calculated to fully suppress the background clutter and noise.And the fusion feature mapping corresponding to the image was formed by combining saliency mapping and multi-scale difference histogram.Finally,the adaptive threshold segmentation method was introduced to accurately locate the real target from the fusion feature map.Multiple sets of test data show that,compared with the existing infrared target detection technology,the proposed algorithm can better locate small and weak targets,and have an ideal ROC curve.