Infrared small target detection is an important research topic in the field of infrared image processing and has a major impact on applications in areas such as remote sensing, infrared imaging precise. Due to atmospheric scattering, refraction and the effect of the lens, the infrared detector to receive the target information very weak, it’s difficult to detect the small target in complex background. In this paper, a novel small target detection method in a single infrared image is proposed based on deep convolutional neural network that is mainly using to extract the features of target, through the method can obtain more discriminative features of infrared image. Firstly, the off-line training of convolution kernel parameters using open data sets and simulated data sets, the result of preliminary training gives an initial convolution kernel, this step can reduce the time required for parameter training. Secondly, the input infrared image is preliminarily processed by the trained parameters to obtain the primary features of the infrared image, through the processing of the convolution kernel, a large number of feature information in different scales of the input image are obtained. Finally, selecting and merging the features, design the efficient characteristic information selection strategy, then fine-tune the convolution parameters with the result information, by merging the feature graph can realize the output of the result target image. The experimental results demonstrated that compared with existing classical methods, the proposed method could greatly improve the quality of the results, more importantly, our method can directly achieve the end-to-end mapping between the input images and target detection results.
In the processing of infrared small target image which has low signal-to-noise ratio and complex background, the target detection and recognition are very hard. So, how to suppress infrared complex background in low signal-to-clutter addition becomes the key problem in the detection of infrared small target image. The topological derivative can quantify the sensitivity of a problem when the domain under consideration is perturbed by changing its topology. Considering the idea of topology optimization, a modified topological derivative based background suppression method for infrared dim small target detection was proposed. An appropriate functional and variational problem is related to the cost function. Thus, the corresponding topological derivative can be used as an indicator function leads to the processed image through a minimization process. Firstly, introduce perturbations to each pixel of the infrared image. Secondly, calculate the corresponding topological derivative. These pixels also have the least cost function. Finally, using the modified optimal diffusion coefficient to diffuse the pixels where the topological derivative is negative to make its background smooth and achieve the purpose of removing the background clutter while enhancing the small target. Compared with other several experiment results of existing background suppressing methods in indexes, the method the paper proposed has innovative ideas and gets well effects of background suppressing and are practical methods. All of above have the important research value for the related work in future.
High dynamic range infrared image detail enhancement is an important processing procedure in the field of infrared (IR) imaging. Because of the dynamic range of natural scene image far beyond the human vision system, display equipment, and the high dynamic image transformed directly from high dynamic to low dynamic will cause detail information lost, it is essential to compress dynamic range of image and enhance detail. Aiming at the disadvantages of existing methods, high dynamic infrared image compressive enhancement based on fast local Laplacian filters were proposed. First, the fast local Laplacian filters are utilized to separate the image into a base layer and detail layer. Second, the dynamic range of base layer was compressed by using gamma correction in order to improve contrast. The detail layer is stretched by utilizing sigmoid function. Finally, the enhanced output image is obtained by recombining the detail layer and base layer. Compared with other methods such as histogram equalization, bilateral filtering, the experimental results demonstrated that the proposed method have a better performance in term of enhancing details and improving contrast by using evaluation index of image detail enhancement.