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]

In this paper, a new method for image mosaicing (image stitching) is introduced based on Scale Invariant Feature transform (SIFT). One of the main drawbacks of SIFT is the redundancy of the extracted keypoints, which leads to lower image mosaicing quality. Recently, a new method called Redundant Keypoint Elimination (RKEM) was presented to remove these redundant features, and enhance image registration performance. Despite the applicability of RKEM, its threshold value is considered the same in all parts of the image. This characteristic leads to inappropriate removal of keypoints due to the fact that distribution of keypoints in the high-detailed region is denser than the low-detailed ones. This paper proposes a new method to improve RKEM called Clustered RKEM (CRKEM) which is based on keypoints distribution. Moreover, in this paper a new blending algorithm is proposed based on a Gaussian-weighted function. In the proposed blending method, the Gaussian function is proposed based on the mean and variance of the pixels in the overlapped region of images to be mosiaced. In comparison with the classical methods, the experimental results confirm the superiority of the proposed method in image mosaicing as well as to image registration and matching. [ABSTRACT FROM AUTHOR]

Registration between remote sensing images has been a research focus in the field of remote sensing image processing. Most of the existing image registration algorithms applied to feature point matching are derived from image feature extraction methods, such as scale-invariant feature transform (SIFT), speed-up robust features (SURF) and Siamese neural network. Such methods encounter difficulties in achieving accurate image registration where there is a large bias in the image features or no significant feature points. Aiming to solve this problem, this paper proposes an algorithm for multi-source image registration based on geographical location information (GLI). By calculating the geographic location information that corresponds to the pixel in the image, the ideal projected pixel position of the corresponding image is obtained using spatial coordinate transformation. Additionally, the corresponding relationship between the two images is calculated by combining multiple sets of registration points. The simulation experiment illustrates that, under selected common simulation parameters, the average value of the relative registration-point error between the two images is 12.64 pixels, and the registration accuracy of the corresponding ground registration point is higher than 6.5 m. In the registration experiment involving remote sensing images from different sources, the average registration pixel error of this algorithm is 20.92 pixels, and the registration error of the image center is 21.24 pixels. In comparison, the image center registration error given by the convolutional neural network (CNN) is 142.35 pixels after the registration error is manually eliminated. For the registration of homologous and featureless remote sensing images, the SIFT algorithm can only offer one set of registration points for the correct region, and the neural network cannot achieve accurate registration results. The registration accuracy of the presented algorithm is 7.2 pixels, corresponding to a ground registration accuracy of 4.32 m and achieving more accurate registration between featureless images. [ABSTRACT FROM AUTHOR]

This paper presents a 3D reconstruction method for fast elevation determination on construction sites. The proposed method is intended to automatically and accurately determine construction site elevations using drone-based, low–high orthoimage pairs. This method requires fewer images than other methods for covering a large target area of a construction site. An up–forward–down path was designed to capture approximately 2∶1 -scale images at different altitudes over target stations. A pixel grid matching and elevation determination algorithm was developed to automatically match images in dense pixel grid-style via self-adaptive patch feature descriptors, and simultaneously determine elevations based on a virtual elevation model. The 3D reconstruction results were an elevation map and an orthoimage at each station. Then, the large-scale results of the entire site were easily stitched from adjacent results with narrow overlaps. Moreover, results alignment was automatically performed via the U-net detected ground control point. Experiments validated that in 10–20 and 20–40 orthoimage pairs, 92% of 2,500- and 4,761-pixels were matched in the strongest and strong levels, which was better than sparse reconstructions via structure from motion; moreover, the elevation measurements were as accurate as photogrammetry using multiscale overlapping images. [ABSTRACT FROM AUTHOR]

Dense image matching (DIM) is a critical technique when computing accurate 3D geometric information for many photogrammetric applications. Most DIM methods adopt first‐order regularisation priors for efficient matching, which often introduce stepped biases (also called fronto‐parallel biases) into the matching results. To remove these biases and compute more accurate matching results, this paper proposes a novel post‐filtering method by adjusting the surface orientation of each pixel in the matching process. The core algorithm formulates the post‐filtering as the optimisation of a global energy function with second‐order regularisation priors. A compromise solution of the energy function is computed by breaking the optimisation into a collection of sub‐optimisations of each pixel in a local adaptive window. The proposed method was compared with several state‐of‐the‐art post‐filtering methods on indoor, aerial and satellite datasets. The comparisons demonstrate that the proposed method obtains the highest post‐filtering accuracies on all datasets. [ABSTRACT FROM AUTHOR]

Image matching plays an important role in Augmented Reality, Simultaneous Localization and Map ping (SLAM), and unmanned. The key issue of image matching is the accuracy of feature matching between adjacent images. Due to the complicated environment, matching between images through geometry feature runs into bottlenecks. In this paper, we focus on improving the accuracy of feature matching by incorporating instance-aware semantic segmentation into Oriented FAST and Rotated BRIEF (ORB) feature matching, which is broadly utilized in image registration and Visual SLAM. We segment the objects between the adjacent images on pixel level and qualify the features matching procedure using the semantic information. Experiment results show that our novel proposed method increases the accuracy of feature matching. [ABSTRACT FROM AUTHOR]

We propose a fast elemental image generation algorithm for integral imaging display based on depth information and a two-dimensional scene image. Existing elemental image generation algorithms establish many-to-one mappings from the scene to the elemental image, thereby including redundant calculations. The improved algorithm entails establishing a single mapping from elemental image to the scene. Each pixel in the elemental image matches a unique scene pixel by inverse iteration along the optical path. This establishes a high precision matching and eliminates hole caused by depth discontinuities. In this paper, we obtain a unique mapping solution by inverse iteration along the optical path using a one-pixel step. The time complexity of this algorithm is mainly driven by the pixel sum of the elemental image array. We generate the elemental image seven times faster than the original algorithm. The larger total quantity of the scene pixels further demonstrates the advantages of our algorithm. The experiments show that the update rate can reach more than 30 fps according to the VCD standard (352×240 resolution) using four parallel calculation threads with a four-core Intel® Core(TM) i7-7500U 2.70 GHz CPU despite various delay factors. The results verify the correctness of the theoretical analysis. [ABSTRACT FROM AUTHOR]

To meet the high frame rate requirements of correct point correspondences with a sub-pixel precision, this paper first proposes a Field Programmable Gate Array (FPGA) architecture that consists of corner detection, corner matching, outlier rejection, and sub-pixel precision localisation. In the architecture, a combined Features from Accelerated Segment Test (FAST)+ Binary Robust Independent Elementary Features (BRIEF) algorithm is adopted for detection and matching with pixel precision, a combined algorithm of Slope-based Rejection (SR) and Correlation-Coefficient-based Rejection (CCR) is used to reject the outliers, and a gradient centroid-based algorithm is used for sub-pixel precision localisation. The whole FPGA architecture is implemented on a single FPGA platform (Xilinx XC72K325T). Five image datasets with different spatial resolutions, textures, lights, rotate angles, and viewpoints are used to evaluate the performance of the FPGA-based implementation. The experimental results show that (1) the SR and CCR algorithms are effective for outlier rejection; (2) a higher correct matching rate is achieved for the image pairs that cover artificial textures than for those that cover natural textures; (3) the proposed architecture is also suitable for image pairs with small change of lights, rotate angles, and viewpoints; (4) the speed of the FPGA-based implementation can reach 280 Frames Per Second (FPS), which is 35 times faster than the Personal Computer (PC)-based implementation; and (5) the usage of FPGA resources is acceptable for the selected FPGA platform. The speed and usage of the FPGA resources can be improved when the whole FPGA-based implementation is further optimised. [ABSTRACT FROM AUTHOR]

Summary: This paper proposes an all‐hardware architecture to perform the subpixel refinement operation in the scale invariant transform algorithm. Although the literature describes several hardware implementations of this algorithm, due to its complexity, most of them are based on simplifications of it. These implementations normally exclude the subpixel refinement stage, which, however, is an essential process to obtain accurate results in image matching applications.The architecture has been described in very high–speed integrated circuit hardware description language at register transfer level and synthesized on a Xilinx Zynq 7020 device. The latency of the proposed architecture to generate a refinement operation is 211 clock cycles, and the throughput achieved exploiting pipeline techniques is 64 cycles. The architecture uses fixed‐point data representation and has been tested with images from known databases, yielding very good performance compared with the floating‐point software implementation of the algorithm. [ABSTRACT FROM AUTHOR]

This paper presents an improved Oriented Features from Accelerated Segment Test (FAST) and Rotated BRIEF (ORB) algorithm named ORB using three-patch and local gray difference (ORB-TPLGD). ORB takes a breakthrough in real-time aspect. However, subtle changes of the image may greatly affect its final binary description. In this paper, the feature description generation is focused. On one hand, instead of pixel patch pairs comparison method used in present ORB algorithm, a three-pixel patch group comparison method is adopted to generate the binary string. In each group, the gray value of the main patch is compared with that of the other two companion patches to determine the corresponding bit of the binary description. On the other hand, the present ORB algorithm simply uses the gray size comparison between pixel patch pairs, while ignoring the information of the gray difference value. In this paper, another binary string based on the gray difference information mentioned above is generated. Finally, the feature fusion method is adopted to combine the binary strings generated in the above two steps to generate a new feature description. Experiment results indicate that our improved ORB algorithm can achieve greater performance than ORB and some other related algorithms. [ABSTRACT FROM AUTHOR]