Narrow Results

A novel image passive forensics method for copy-move forgery detection is proposed. The proposed method combines block matching technology and feature point matching technology, and breaks away from the general framework of the visual feature-based approach that used local visual feature such as SIFT and followed by a clustering procedure to group feature points that are spatially close. In our work, image keypoints are extracted using Harris detector, and the statistical features of keypoint neighborhoods are used to generate forensics features. Then we proposed a new forensics features matching approach, in which, a region growth technology and a mismatch checking approach are developed to reduce mismatched keypoints and improve detected accuracy. We also develop a duplicate region detection method based on the distance frequency of corresponding keypoint pairs. The proposed method can detect duplicate regions for high resolution images. It has higher detection accuracy and computation efficiency. Experimental results show that the proposed method is robust for content-preserving manipulations such as JPEG compression, gamma adjustment, filtering, luminance enhancement, blurring, etc.

Extraction of agricultural parcels from high-resolution satellite imagery is an important task in precision agriculture. Here, we present a semi-automatic approach for agricultural parcel detection that achieves high accuracy and efficiency. Unlike the techniques presented in previous literatures, this method is pixel based, and it exploits the properties of a spectral angle mapper (SAM) to develop customized operators to accurately derive the parcels. The main steps of the method are sample selection, textural analysis, spectral homogenization, SAM, thresholding, and region growth. We have systematically evaluated the algorithm proposed on a variety of images from Gaofen-1 wide field of view (GF-1 WFV), Resource 1-02C (ZY1-02C), and Gaofen-2 (GF-2) to aerial image; the accuracies are 99.09% of GF-1 WFV, 84.42% of ZY1-02C, 96.51% and 92.18% of GF-2, and close to 100% of aerial image; these results demonstrated its accuracy and robustness.

LSS-Target (the Low altitude, Slow speed and Small Target) is likely to be a threat to the observation platform, thus infrared LSS-Target detection is an urgent task. LSS-Target is a challenging issue due to the low Signal-to-Noise Ratio (SNR) and sophisticated background. Motivated by the analysis of infrared imaging characteristics, this paper proposes a novel fusion method for IR LSS-Target detection with complex urban background, which is suitable for precise guidance and self defense. First, an adaptive threshold segmentation based on accumulative histogram and maximum likelihood estimation are utilized to eliminate the clutter and improve SNR of the initial image. Second, a template is set up to identify the seed points in the image. Third, a constrained four criteria region growth algorithm is performed to separate the entire regions. Finally, the confidence measure is constructed, which can eliminate false targets and the background edges. Experimental results show that the method in this paper can screen out the real LSS-Target in real time with high accuracy under sophisticated background.