Image Registration Using Adaptive Polar Transform

Image registration is an essential step in many image processing applications that need visual information from multiple images for comparison, integration or analysis. Recently researchers have introduced image registration techniques using the log-polar transform (LPT) for its rotation and scale invariant properties. However, the accuracy of the approach limits by the number of samples used in the mapping process, which affects directly the computational cost. Motivated by the success of LPT based approach and its limitation, we propose a novel Projective Polar Transform (PPT) based image registration method that is robust to translation, scale, and rotation and yields high accuracy while requires low computational cost. Unlike LPT that 2D interpolation is needed in the mapping process, our method uses one-to-one mapping mechanism that directly arranges image from Cartesian to Polar coordinate according to pre-computed PPT map. An innovative projection mechanism is proposed to reduce the image from two to one dimensional vectors. With the proposed approach, the mapping procedure is accelerated and the matching process can be performed in 1D. This results in great reduction of the computational cost.

Registration is the process of finding transformations that makes correspondence between related image pairs so that pixels in the two images precisely coincide to the same points in the scene. Once registered, the image can be combined or fused in a way that improves useful information extraction. The log-polar transform (LPT) is a well known space variant image registration scheme used for medical images. However though LPT is invariant to rotation and scale changes, it does not support translation faithfully. In this paper, a hybrid algorithm for image registration that produces faithful mapping under affine geometrical distortion using LPT, Fourier transform (FT), and phase correlation has been presented. The resultant algorithm is invariant to rotation, scale and translation. The proposed algorithm is verified on medical images with affine shifting, partial data, and in the presence of noise.

Under the background of telemedicine, a new registration and mosaic algorithm for medical images is proposed in this paper to solve the problems of electronic noise, uneven illumination and ray scattering in the real-time medical process. The improved Retinex algorithm by trilateral filter and homomorphic filtering algorithm can enhance the image effectively in preprocessing. The improved phase correlation algorithm based on log polar transformation was used to calculate parameters, such as rotation, scaling and translation. Then, the SUSAN corner matching points were extracted in overlapping positions, the improved KD tree was used for enhancing matching efficiency. Later, matching points were purified by the improved RANSAC algorithm. Finally, images were processed by Laplacian pyramid decomposition algorithm to make the image joint seemed smooth and natural. The results of experiments and evaluation criteria confirm that the new method has high robustness in the process of medical image registration and stitching in the network.

The image navigation and registration (INR) performance assessment for the Geostationary Operational Environmental Satellite-R (GOES-R) Advanced Baseline Imager (ABI) is an image registration process at the subpixel level, which involves the registration of images from GOES-R ABI with corresponding images from Landsat at the same scene. We present an optimized image registration algorithm for GOES-R ABI INR performance assessment (GAIPA): a gradient descent algorithm with a Sobel edge-enhancement procedure and a normalized lookup table (NLUT) image transformation (GDSN), which is an area-based image registration approach with Pearson correlation as the similarity metric. Sobel edge enhancements increase the sensitivity of Pearson correlations that lead to much improved registration accuracy. NLUT transformations minimize radiance differences between GOES-R ABI images acquired at different times and Landsat reference images on image registration outcomes, which lead to improvements in correlation values and quality of image registration outputs. Evaluations of registration accuracy and quality metrics in GDSN show considerable improvements in computational efficiency, registration accuracy, and quality over the approach implemented on the image INR performance assessment tool set. Pearson correlations can be effective and robust similarity metrics in image registration at the subpixel level with implementing Sobel edge enhancement and NLUT transformation. The statistical analysis of the image registration outputs for GAIPA shows the new algorithm provides considerable improvements in reducing the uncertainties in the registration outputs.

Image registration technology has been widely used in many parts of the computer vision system such as the automatic optical inspection system which is used to detect the printed circuit board (PCB) defects. The accuracy of the image registration will deeply influence the system's performance, so this study proposed an accurate image registration algorithm and applied it to the PCB defect detection. Good features to track feature detector and speeded up robust feature descriptor are combined to extract efficient features to achieve the first accurate image registration. Afterwards, cross‐correlation functions were used to compute the shift between the reference image and the first‐registered image for further accurate registration. Experimental results show that the authors’ algorithm performs a much better registration, with a lower root‐mean‐square error value between the reference image and transformed image. What is more, they applied it to detect the defects of PCB with a high accuracy.

In this work, a hardware efficient and flexible architecture has been proposed for computing Log Polar Transformation (LPT) for real time applications in image processing and pattern recognition. CORDIC (COordinate Rotational DIgital Computing) is used as a building block for our proposed architecture, so as to keep the design flexible that can be adapted to different specific applications. The basic philosophy of CORDIC algorithm is that it can be implemented using SHIFT and ADD/SUB units using VLSI technologies. Purely pipelined CORDIC architecture has been used to get high throughput. This proposed architecture has been simulated and synthesized using RTL Coding with Verilog-XL and Design Vision. The synthesized result shows that the circuit can be used with maximum frequency of 125 MHz with area 9.5 mm using 0.18 μm CMOS technology.

New technological developments in optics and electronics are rendering laser scanning systems cheaper and more accurate. Such systems can directly capture depth information from objects simplifying range image analysis and enlarging their applications scope. Accurate image registration algorithms represent a pivotal aspect of range image analysis which still require substantial improvements. While a large number of algorithms have been proposed, it is very important to compare the performance of different registration algorithms so that good algorithms can be selected for real applications. Unfortunately, most of these algorithms are iterative with a number of parameters initialised by users. The main disadvantages of these methods are that these parameters are often context dependent. Thus, while aiming to develop automatic registration algorithms instead of comparing the registration algorithms themselves, we compare the structural constraints based on a single registration algorithm(GICP). A large number of experiments using real range images demonstrate their effectiveness on improving the GICP algorithm.

This paper describes a novel image registration method that combines log-polar transform and SIFT to recover similarity transformations (rotation/ scale/ translation). We extracts SIFT feature points in the two images firstly. Then, we use threshold Euclidean distance to coarsely match the feature points. After that, the log-polar transform is applied to compute the rotation and scale parameters. And we can obtain the translation parameter by the location relationship of the feature points.

Feature detection and description is a key concept of image processing and computer vision. Oriented FAST and Rotated BRIEF (ORB) algorithm is one of feature detector and descriptor that is widely used in image stitching and registration systems such as simultaneous localization and mapping (SLAM) systems. ORB algorithm has improved the speed of processing and robustness. It has scale-invariance and rotational invariance properties. However, there are some obstacles to implement on low-power devices such as its high computational intensity. This paper presents the feature detection and description algorithm based on ORB algorithm which adopted parallel and pipeline techniques to accelerate the system and be beneficial for designing FPGA architecture. Compared with the results obtained by cv.ORB() function, the proposed algorithm improves the parallelism and reduces the operation of the process but could maintain the consistency of the corner numbers and the accuracy.

Improved Rapid Corner Detection Algorithm in Medical Image

To balance dose reduction and image registration accuracy in breast setup imaging. In particular, the authors demonstrate the relationship between scan angle and dose delivery for cone beam tomosynthesis (CBTS) when employed for setup verification of breast cancer patients with surgical clips. The dose measurements were performed in a female torso phantom for varying scan angles of CBTS. Setup accuracy was measured using three registration methods: Clip centroid localization accuracy and the accuracy of two semiautomatic registration algorithms. The dose to the organs outside of the ipsilateral breast and registration accuracy information were compared to determine the optimal scan angle for CBTS for breast patient setup verification. Isocenter positions at the center of the patient and at the breast-chest wall interface were considered. Image registration accuracy was within 1 mm for the CBTS scan angles theta above 20 degrees for some scenarios and as large as 80 degrees for the worst case, depending on the imaged breast and registration algorithm. Registration accuracy was highest based on clip centroid localization. For left and right breast imaging with the isocenter at the chest wall, the dose to the contralateral side of the patient was very low (<0.5 cGy) for all scan angles considered. For central isocenter location, the optimal scan angles were 30 degrees - 50 degrees for the left breast imaging and 40 degrees - 50 degrees for the right breast imaging, with the difference due to the geometric asymmetry of the current clinical imaging system. The optimal scan angles for CBTS imaging were found to be between 10 degrees and 50 degrees, depending on the isocenter location and ipsilateral breast. Use of the isocenter at the breast-chest wall locations always resulted in greater accuracy of image registration (<1 mm) at smaller angles (10 degrees - 20 degrees) and at lower doses (<0.1 cGy) to the contralateral organs. For chest wall isocenters, doses delivered to organs outside of the target breast were much smaller than the scattered and leakage doses of the treatment beams. The complete volumetric information of all clips in the region of interest, combined with the small dose to the contralateral organs and the small scan angle, could result in an advantage for small angle CBTS with off center isocenters over simple orthogonal pairs.

Objective To evaluate the image quality and registration accuracy of a three-dimensional (3D) dynamic phantom in four-dimensional computed tomography (4DCT) and four-dimensional cone-beam computed tomography (4DCBCT). Methods The Computerized Imaging Reference Systems Dynamic Thorax Phantom Model 008A was scanned to get 4DCT and 4DCBCT images. Two balls with different diameters (O= 1 cm and O= 2 cm) were used to simulate tumors with different sizes. The motion mode of the balls was 3D sinusoidal motion at 0.25 Hz (the amplitudes along the x, y, and z axes were ±1.0 cm, ±0.4 cm, and ±0.2 cm, respectively). Gross target volumes (GTVs) from 10-phase bins, internal gross target volumes (IGTV), and target volumes on maximum intensity projection (MIP) and mean intensity projection (MeanIP) images were contoured and calculated. Target volumes on 4DCT or 4DCBCT images were compared with the static and dynamic volumes of the balls (VS and VD). The matching index (MI) of target volumes between the 4DCT and 4DCBCT images was analyzed after rigid image registration. Results The GTV in each phase of the image was larger than VS. The difference between the average GTV derived from 10 phases of 4DCT or 4DCBCT images and Vs of the small ball was larger than that of the large ball (35.03% vs. 22.66%; 32.62% vs. 17.00%). All the IGTVs and target volumes on MIP images were slightly larger than VD, but target volumes on MeanIP images were smaller than VD. The average MI of 10-phase bins of the small ball was smaller than that of the large ball (66.76% vs. 82.21%). Moreover, MIs of IGTV, MIP, and MeanIP of the small ball were also smaller than those of the large ball (77.39% vs. 90.29%; 75.90% vs. 89.28%; 74.47% vs. 82.74%). Conclusions In the case of a relatively small tumor volume and a relatively large motion amplitude, 4DCT and 4DCBCT should be used with caution for comparison of image registration. Key words: Tomography, X-ray computed; Tomography, X-ray computed, cone-beam; Imaging registration; Phantom

This thesis focuses on the rigid-body registration of a three-dimensional model of an object to a set of its two-dimensional projections. The main contribution is the development of two registration algorithms that use a continuous model of the volume based on splines, either in the space domain or in the frequency domain. This allows for a well-defined gradient of the dissimilarity measure, which is a necessary condition for efficient and accurate registration. The first part of the thesis contains a review of the literature on volume-to- image registration. Then, we discuss data interpolation in the space domain and in the frequency domain. The basic concepts of our registration strategy are given in the second part of the thesis. We present a novel one-step approach for fast ray casting to simulate space-based volume projections. We also discuss the use of the central-slice theorem to simulate frequency-based volume projections. Then, we consider the question of the registration robustness. To improve the robustness of the space-based approach, we apply a multiresolution optimization strategy where spline-based data pyramids are processed in coarse-to-fine fashion, which improves speed as well. To improve the robustness of the frequency-based registration, we apply a coarse-to-fine strategy that involves weights in the frequency domain. In the third part, we apply our space-based algorithm to computer-assisted orthopedic surgery while adapting it to the perspective projection model. We show that the registration accuracy achieved using the orthopedic data is consistent with the current standards. Then, we apply our frequency-based registration to three-dimensional electron-microscopy application. We show that our algorithm can be used to obtain a refined solution with respect to currently available algorithms. The novelty of our approach is in dealing with a continuous space of geometric parameters, contrary to the standard methods which deal with quantized parameters. We conclude that our continuous parameter space leads to better registration accuracy. Last, we compare the performance of the frequency-based algorithm with that of the space-based algorithm in the context of electron microscopy. With these data, we observe that frequency-based registration algorithm outperforms the space-based one, which we attribute to the suitability of interpolation in the frequency domain when dealing with strictly space-limited data.

Image registration is a digital image processing technique that takes two or more of images of a scene in different coordinate systems and transforms them into a single coordinate system. Image registration is a necessary step in many advanced image processing techniques, such as multi-frame super-resolution. For that reason, registration accuracy is very crucial. While image registration is usually performed on images, one can perform the registration using metric images as well. This paper will present registration methods and their accuracies for various noise levels for the case of pure translational image motion. Registration techniques will be applied to the images themselves as well as to phase congruency images, gradient images, and edge-detected images. This study will also investigate registration of under-sampled images. Noise-free images are degraded using three types of noise: additive Gaussian noise, fixed-pattern noise along the column direction, and a combination of these two. The registration error is quantified for two registration algorithms with three different images as a function of the signal-to-noise ratio. A test on the usefulness of the image registration and registration accuracy performed on the intensity images of the Stokes imaging polarimeter. The Stokes images calculated before and after registration of the intensity images are compared to each other to show the improvement.

Pressure-sensitive paint technology is a wind tunnel pressure measurement frontier technology with high economical efficiency and high speed. In the wind tunnel test, due to the strong wind, the model will be distorted, making the wind image and the windless image difficult to register, which will seriously affect the pressure measurement accuracy. In response to this problem, this paper innovatively applies the two-dimensional non-rigid iterative closest point (ICP) algorithm to solve this problem. The point cloud method is used to make the image detail area to be effectively registered, and it is also conducive to the subsequent three-dimensional reconstruction work. However, due to the two-dimensional non-rigid ICP algorithm, only the two-dimensional coordinate positional relationship is considered. The correlation of the pixel grayscales of the pressure-sensitive paint image is neglected, so that the registration accuracy is not too high. However, if the three-dimensional non-rigid ICP algorithm is directly used, misregistration will occur. Therefore, in order to further improve the registration accuracy, this paper proposes a non-rigid ICP algorithm based on pixel-based search strategy. The algorithm designs a dual-target search strategy that takes 2D coordinates and pixel gray values into consideration and achieves accurate local matching, realizing point search and double goal optimization. The algorithm of this paper is compared with five registration algorithms on multiple sets of pressure sensitive paint images. The experimental results show that the proposed algorithm has the best registration accuracy. Compared to the suboptimal algorithm, the RMSE is improved by more than 15% and the NMI is increased by about 5%.