Search

The method of detecting the whole tampered image increases many unnecessary calculations. In order to reduce the computational complexity and further improve the detection accuracy, this paper proposed a copy-move forgery detection method based on improved saliency map and local feature matching. Firstly, it combined the gradient of image to improve the saliency map, and separated the local salient regions containing high texture information of the image. Secondly, it only used SIFT（Scale Invariant Feature Transform） algorithm to extract feature points in this local area. Then, it adopted density clustering and two-stage matching strategy for images with low saliency, and adopted the strategy of superpixel segmentation and salient block feature matching for images with high saliency. Finally, it combined PSNR and morphological operations to locate the tampered area. Experiments on two public datasets show that the average detection time of this method is less than 10s, and the average detection accuracy is greater than 97%,which are better than the compared methods. The experimental results show that this method can greatly reduce the detection time, effectively improve the detection accuracy, and has good robustness to geometric transformation and post-processing operations. [ABSTRACT FROM AUTHOR]

Aiming at the problems of low efficiency and accuracy of traditional artificial pavement crack detection, as well as a lot of manpower and material resources, a pavement crack detection method based on fractional domain spectrum windowing and contrast enhancement was proposed. Firstly, the fractional Fourier transform was used to transform the crack image into the fractional domain, and the optimal window function was used to window the spectrum to achieve the denoising effect. Then, the fractional-order homomorphic filtering algorithm was used to enhance the contrast of the denoised image, and the image with the highest contrast under the best order was obtained through an iterative method. Finally, the Canny algorithm and the Otsu algorithm were used to perform edge detection and threshold segmentation on the enhanced image, respectively, and to effectively extract the crack features in the crack image by performing the OR operation and the image morphology operation on the two results. Comparing the detection results with the fractional frequency domain processing method and the improved histogram contrast (HC) saliency method, the accuracy of this method is increased by 5.84% and 4.5%, and the recall rate is increased by 5.58% and 3.52%, respectively, indicating that the method has advantages in road crack detection, better detection effect and higher recognition rate. [ABSTRACT FROM AUTHOR]