Proposed Image Registration Method Research Articles

Developing thermal infrared de-ghost and multi-level nested conglutinated segmentation algorithm for detection of rice seed setting rate

The seed setting rate (SSR) of rice is not only a key component of yield, but also an important parameter in rice phenotypic analysis. Fast and accurate detection of SSR is of great significance for yield prediction. The purpose of this research was to detect the SSR of rice quickly and automatically. A thermal infrared–visible light dual imaging system was built to obtain thermal infrared images and RGB images of rice grains. This paper proposed image registration method, thermal infrared de-ghost method and multi-layer nested conglutinated segmentation algorithm to detect SSR. Compared with the detection accuracy of three deep learning models (Faster RCNN 96.43%, SSD 81.84%, YOLO V3 96.75%) and image registration methods (80.83%), the highest SSR detection accuracy (97.66%) was achieved by fusing thermal infrared de-ghost and multi-layer nested conglutinated segmentation algorithm. This method has the advantages of simple structure, high efficiency and competitive results, and has great potential in detecting seed setting rate.

Utilizing Mask RCNN for Monitoring Postoperative Free Flap: Circulatory Compromise Detection Based on Visible-Light and Infrared Images

The new postoperative free flap monitoring system combines visible-light and infrared techniques to overcome the limitations of our previous study, such as resistance to illumination change, patients with large movements, and the infrared images being too blurry to identify the boundary. In the visible-light system, the Mask region-based convolutional neural networks (RCNN) was adopted to segment the region of the free flap, and these time-course visible images were aligned using our proposed image registration method. Then, the registered visible images were projected onto the infrared image by the coordinate transformation. The analysis method adopted the residual factor analysis to extract the more unvarnished specific factors. The experiments were divided into two parts. In image processing, the accuracy of the coordinate transformation has a mean error of 1.78 pixels and a standard deviation of 0.98 pixels. The segmentation results showed excellent performance in the most severe case with apparent motion and rotation, covered by gauze and the ventilator, as well as illumination variation. The dice coefficient is 0.9551 ± 0.0158, and the Hausdorff distance is 2.3943 ± 0.3921 pixels. The image registration results also reveal that the Canny edge of the deformed image was superimposed onto the reference image well. In circulatory compromise detection, the vascular congestion was detected much earlier than manual observation, and the classified type of occlusion was the same as the clinical reports. Therefore, the dual-camera monitoring system provides a reliable tool for the surgeon to hold onto the chance of repairing the free flap with vascular obstruction.

Faster dense deformable image registration by utilizing both CPU and GPU.

.Purpose: Image registration is an important aspect of medical image analysis and a key component in many analysis concepts. Applications include fusion of multimodal images, multi-atlas segmentation, and whole-body analysis. Deformable image registration is often computationally expensive, and the need for efficient registration methods is highlighted by the emergence of large-scale image databases, e.g., the UK Biobank, providing imaging from 100,000 participants.Approach: We present a heterogeneous computing approach, utilizing both the CPU and the graphics processing unit (GPU), to accelerate a previously proposed image registration method. The parallelizable task of computing the matching criterion is offloaded to the GPU, where it can be computed efficiently, while the more complex optimization task is performed on the CPU. To lessen the impact of data synchronization between the CPU and GPU, we propose a pipeline model, effectively overlapping computational tasks with data synchronization. The performance is evaluated on a brain labeling task and compared with a CPU implementation of the same method and the popular advanced normalization tools (ANTs) software.Results: The proposed method presents a speed-up by factors of 4 and 8 against the CPU implementation and the ANTs software, respectively. A significant improvement in labeling quality was also observed, with measured mean Dice overlaps of 0.712 and 0.701 for our method and ANTs, respectively.Conclusions: We showed that the proposed method compares favorably to the ANTs software yielding both a significant speed-up and an improvement in labeling quality. The registration method together with the proposed parallelization strategy is implemented as an open-source software package, deform.

REGISTRATION OF OPTICAL AND SAR SATELLITE IMAGES BASED ON GEOMETRIC FEATURE TEMPLATES

Abstract. Image registration is required for different remote sensing applications, like change detection or image fusion. Since research studies have shown the outstanding absolute geometric accuracy of high resolution radar satellites images like TerraSAR-X, the importance of SAR images as source for geolocation enhancement has increased. Due to this fact, multi-sensor image to image registration of optical and SAR images can be used for the improvement of the absolute geometric processing and accuracy of optical images with TerraSAR-X as reference. In comparison to the common optical and SAR image registration methods the proposed method is a combination of intensity-based and feature-based approaches. The proposed method avoids the direct and often difficult detection of features from the SAR images. SAR-like templates are generated from features detected from the optical image. These templates are used for an intensity-based matching with the SAR image. The results of the matching process are ground control points, which are used for the estimation of translation parameters followed by a subpixel translation of the optical image. The proposed image registration method is tested for two pairs of TerraSAR-X and QuickBird images and one pair of TerraSAR-X andWorldView-2 images of a suburban area. The results show that with the proposed method the geometric accuracy of optical images can be enhanced.

Distinctive local binary pattern for non‐rigid registration of lung computed tomography images

Non-rigid registration of lung computed tomography (CT) images is a valuable tool for various clinical applications. Many methods such as bilateral filters and census transform have been used to deal with discontinuity of lung motion and local intensity variation. However, census transform cannot distinguish between low and high contrast regions, which may lead to negative influence to differential-based registration methods. A novel distinctive local binary pattern that can generate distinctive representations of high contrast images is proposed. Combing the novel local binary pattern, bilateral filters, the inverse-consistent symmetrical method and the Lucas–Kanade method, a novel accurate image registration method is developed. The experiments are performed on the publicly available 4D CT lung dataset from DIR-Lab. Compared with the census transform, the proposed distinctive local binary pattern can achieve relatively better results. The proposed image registration method greatly improves the accuracy of the classical Lucas–Kanade method and the bilateral filters-based Demons. In addition, the proposed registration method is most accurate among all unmasked methods tested on this dataset.

Key technology of D-InSAR at X-band for monitoring land subsidence in mining area and its application

In theory, land subsidence measurement results with high accuracy can be obtained by using the Differential Interferometry Synthetic Aperture Radar (D-InSAR) at X-band. In practice, however, the measuring accuracy of D-InSAR at X-band has been seriously affected by some factors, e.g., decorrelation and high deformation gradient. In this work, the monitoring capability of D-InSAR for coal-mining subsidence is evaluated by using SAR data acquired by TerrraSAR-X system. The SAR image registration method for low coherence image pairs, the denoising phase filter for high noise level interferogram and atmospheric effects mitigation method are the key technical aspects which directly influence the measurement results of D-InSAR at X-band. Thus, a robust image registration method, an improved phase filter method and an atmospheric effects mitigation method are proposed in this paper. The proposed image registration method successfully achieves InSAR coregistration, while the amplitude cross-correlation cannot properly coregister low coherence SAR image pairs. Moreover, the time complexity of the proposed image registration method is obviously slighter than that of the Singular Value Decomposition (SVD) method. The comparing experiment results and the unwrapping phase results show that the improved Goldstein filter is more effective than the original Goldstein filter in noise elimination. The atmospheric influence correction experiment results show that the land subsidence areas with atmospheric influence correction are more clarified than that of without atmospheric influence correction. In summary, the presented methods directly improved the measurement results of D-InSAR at X-band.

Adaptive Image Registration via Hierarchical Voronoi Subdivision

Advances in image acquisition systems have made it possible to capture high-resolution images of a scene, recording considerable scene details. With increased resolution comes increased image size and geometric difference between multiview images, complicating image registration. Through Voronoi subdivision, we subdivide large images into small corresponding regions, and by registering small regions, we register the images in a piecewise manner. Image subdivision reduces the geometric difference between regions that are registered and simplifies the correspondence process. The proposed method is a hierarchical one. While previous methods use the same block size and shape at a hierarchy, the proposed method adapts the block size and shape to the local image details and geometric difference between the images. This adaptation makes it possible to keep geometric difference between corresponding regions small and simplifies the correspondence process. Implementational details of the proposed image registration method are provided, and experimental results on various types of images are presented and analyzed.

Multi-resolution remote sensing image registration using differential evolution with adaptive strategy selection

An algorithm based on differential evolution (DE) with adaptive strategy selection for remote sensing image registration is designed. The paper has two major contributions. One contribution is the reduction of the computation time for a better result, i.e., fitness function approximate, which is based on multi-resolution processing and histogram matching. The second contribution is an adaptive strategy known as probability matching integration. The probability matching technique is employed in DE to autonomously select the most suitable strategy while solving the problem. To the best of our knowledge, the proposed method is the first adaptive strategy selection-based approach for image registration. In our experiments, the proposed method significantly outperforms previously proposed image registration method based on DE.

A New Cross-Correlation Based Image Registration Method

Image registration is a fundamental task for combining two or more images taken from different viewpoints, different times, or different sensors. It is a process of determining the point by point correspondence between two images from the same scene. The proposed image registration method uses the area-based approach to process image registration and the objective is to find the maximum similarity through the cross-correlation measure. Most cross-correlation methods are developed based on image intensities for the direct matching purpose. However, it is extremely sensitive to the intensity changes. To counteract illumination effect, the proposed method replaces the intensity with the gradient information, and this concept comes originally from the Hough transform that points having the same parameters and should be on the same line. These two parameters are combined as the similarity between images for image registration. The experimental results obtained by means of several test images illustrate the effectiveness of the proposed image registration method.

An Image Registration Method Based upon Information Theorem on Overlapped Region

Digital image and video have been widely applied to many practical applications due to their simple image acquirement. Image registration is an important image processing for integrating information from images. For Image registration, it is intuitive to orientate images by matching corresponding pixels being considered idealistically identical on the overlapping region. Based on this idea, this article proposes an image registration method that applies the information theorem to the corresponding intensity data. An entropy-based objective function is developed upon the histogram of the intensity differences as to evaluate the similarity between images. Intensity differences represent the differences of the corresponding pixels between the referenced and sensed images on the overlapped region. The sensed image is aligned to the referenced image by minimizing the proposed objective function through iteratively updating the parameters of the projective transformation during the optimization process. The experimental results obtained by means of several test image sets illustrate the effectiveness and feasibility of the proposed image registration method.

Automatic Image Registration Using Improved LBG Algorithm

Image registration is the process of transforming different sets of data into one coordinate system. This paper proposed automatic image registration using improved LBG algorithm. LBG Algorithm used k-means to refine these prototypes, this paper proposed improved k-means algorithm to enhance LBG algorithm. Experimental result shows that the proposed image registration method can improve the precise of image registration, is effective.

Remote Sensing Image Registration Based on Retrofitted SURF Algorithm and Trajectories Generated From Lissajous Figures

In this letter, we propose a novel remote sensing image registration method by optimizing the Speeded Up Robust Features (SURF) and developing a new similarity measure function based on trajectories generated from Lissajous figures. Compared with SURF which has a low feature-matching rate in some complex cases, the retrofitted SURF algorithm is more robust and accurate. The algorithm greatly improves the correct matching rate to over 80%. Furthermore, the recognition capability of the similarity measure is enhanced by using a trajectory disturbance strategy, which is a significant displacement in the trajectory induced by a minor error of the transformation parameters. Experiments show the promising performance of the proposed image registration method.

Image Registration Using Clustering Algorithm

This paper proposed a new method of image registration based on clustering algorithm. It used clustering algorithm to cluster all the feature vectors of images, and adopted EM algorithm to optimize the parameters and algorithm. Experimental result shows that the proposed image registration method can improve the precise of image registration, and reduce error.

Image registration with auto‐mapped control volumes

Many image registration algorithms rely on the use of homologous control points on the two input image sets to be registered. In reality, the interactive identification of the control points on both images is tedious, difficult, and often a source of error. We propose a two-step algorithm to automatically identify homologous regions that are used as a priori information during the image registration procedure. First, a number of small control volumes having distinct anatomical features are identified on the model image in a somewhat arbitrary fashion. Instead of attempting to find their correspondences in the reference image through user interaction, in the proposed method, each of the control regions is mapped to the corresponding part of the reference image by using an automated image registration algorithm. A normalized cross-correlation (NCC) function or mutual information was used as the auto-mapping metric and a limited memory Broyden-Fletcher-Goldfarb-Shanno algorithm (L-BFGS) was employed to optimize the function to find the optimal mapping. For rigid registration, the transformation parameters of the system are obtained by averaging that derived from the individual control volumes. In our deformable calculation, the mapped control volumes are treated as the nodes or control points with known positions on the two images. If the number of control volumes is not enough to cover the whole image to be registered, additional nodes are placed on the model image and then located on the reference image in a manner similar to the conventional BSpline deformable calculation. For deformable registration, the established correspondence by the auto-mapped control volumes provides valuable guidance for the registration calculation and greatly reduces the dimensionality of the problem. The performance of the two-step registrations was applied to three rigid registration cases (two PET-CT registrations and a brain MRI-CT registration) and one deformable registration of inhale and exhale phases of a lung 4D CT. Algorithm convergence was confirmed by starting the registration calculations from a large number of initial transformation parameters. An accuracy of approximately 2 mm was achieved for both deformable and rigid registration. The proposed image registration method greatly reduces the complexity involved in the determination of homologous control points and allows us to minimize the subjectivity and uncertainty associated with the current manual interactive approach. Patient studies have indicated that the two-step registration technique is fast, reliable, and provides a valuable tool to facilitate both rigid and nonrigid image registrations.

Motion tracking in MR-guided liver therapy by using navigator echoes and projection profile matching 1

Rationale and Objectives. Image registration in magnetic resonance (MR) image-guided liver therapy enhances surgical guidance by fusing preoperative multimodality images with intraoperative images, or by fusing intramodality images to correlate serial intraoperative images to monitor the effect of therapy. The objective of this paper is to describe the application of navigator echo and projection profile matching to fast two-dimensional image registration for MR-guided liver therapy. Materials and Methods. We obtain navigator echoes along the read-out and phase-encoding directions by using modified gradient echo imaging. This registration is made possible by masking out the liver profile from the image and performing profile matching with cross-correlation or mutual information as similarity measures. The set of experiments include a phantom study with a 2.0-T experimental MR scanner, and a volunteer and a clinical study with a 0.5-T open-configuration MR scanner, and these evaluate the accuracy and effectiveness of this method for liver therapy. Results. Both the phantom and volunteer study indicate that this method can perform registration in 34 ms with root-mean-square error of 1.6 mm when the given misalignment of a liver is 30 mm. The clinical studies demonstrate that the method can track liver motion of up to approximately 40 mm. Matching profiles with cross-correlation information perform better than with mutual information in terms of robustness and speed. Conclusion. The proposed image registration method has potential clinical impact on and advantages for MR-guided liver therapy.