• Journal of the Semiconductor & Display Technology
• /
• v.18 no.1
• /
• pp.102-104
• /
• 2019

Registering different kinds of clinical images widely used in diagnostic and surgery planning. However, cause of tumor growth or effected by gravity, human tissue has plenty of non-rigid deformation with clinically. Non-rigid registration allows the mapping of straight lines to curves. Therefore, such local deformation makes registration more complicated. In this work, we mainly introduce intra-subject, inter-modality registration. This paper mainly studies the nonlinear registration method of 2D medical image registration. The general medical image registration algorithm requires manual intervention, and cost long registration time. In our work to reduce the registration time in rough registration step, the barycenter and the direction of main axis of the image is calculated, which reduces the calculation amount compared with the method of using mutual information.

• The Journal of Korean Academy of Prosthodontics
• /
• v.59 no.3
• /
• pp.299-304
• /
• 2021

Purpose. The purpose of the present study was to investigate the effects of image matching experience of operators on the accuracy and working time of image registration between radiographic and optical scan images. Materials and methods. Computed tomography and optical scan of a dentate dental arch were obtained. Image matching between the computed tomography and the optical scan (IDC S1, Amann Girrbach, Koblah, Austria) was performed using the point-based automatic registration method in planning software programs (Implant Studio, 3Shape, Copenhagen, Denmark) using two different experience conditions on image registration: experienced group and inexperienced group (n = 15 per group, N = 30). The accuracy of image registration in each group was evaluated by measuring linear discrepancies between matched images, and working time was recorded. Independent t test was used to statistically analyze the result data (α = .05). Results. In the linear deviation, no statistically significant difference was found between the experienced and inexperienced groups. Meanwhile, the working time for image registration was significantly shorter in the experienced group than in the inexperienced group (P = .007). Conclusion. Difference in the image matching experience may not influence the accuracy of image registration of optical scan to computed tomography when the point-based automatic registration was used, but affect the working time for the image registration.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1707-1720
• /
• 2023

With the increase in satellite time series data, the utility of remote sensing data is growing. In the analysis of time series data, the relative positional accuracy between images has a significant impact on the results, making image registration essential for correction. In recent years, research on image registration has been increasing by applying deep learning, which outperforms existing image registration algorithms. To train deep learning-based registration models, a large number of image pairs are required. Additionally, creating a correlation map between the data of existing deep learning models and applying additional computations to extract registration points is inefficient. To overcome these drawbacks, this study developed a data augmentation technique for training image registration models and applied it to OffsetNet, a registration model that predicts the offset amount itself, to perform image registration for KOMSAT-2, -3, and -3A. The results of the model training showed that OffsetNet accurately predicted the offset amount for the test data, enabling effective registration of the master and slave images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.180-194
• /
• 2021

For small image registration, feature-based approaches are likely to fail as feature detectors cannot detect enough feature points from low-resolution images. The classic FFT approach's prediction accuracy is high, but the registration time can be relatively long, about several seconds to register one image pair. To achieve real-time and high-precision rigid registration for small images, we apply deep neural networks for supervised rigid transformation prediction, which directly predicts the transformation parameters. We train deep registration models with rigidly transformed CIFAR-10 images and STL-10 images, and evaluate the generalization ability of deep registration models with transformed CIFAR-10 images, STL-10 images, and randomly generated images. Experimental results show that the deep registration models we propose can achieve comparable accuracy to the classic FFT approach for small CIFAR-10 images (32×32) and our LSTM registration model takes less than 1ms to register one pair of images. For moderate size STL-10 images (96×96), FFT significantly outperforms deep registration models in terms of accuracy but is also considerably slower. Our results suggest that deep registration models have competitive advantages over conventional approaches, at least for small images.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.594-597
• /
• 2006

Image co-registration is the process of overlaying two images of the same scene. One of which is a reference image, while the other (sensed image) is geometrically transformed to the one. Numerous methods were developed for the automated image co-registration and it is known as a time-consuming and/or computation-intensive procedure. In order to improve efficiency and effectiveness of the co-registration of satellite imagery, this paper proposes a pre-qualified area matching, which is composed of feature extraction with Laplacian filter and area matching algorithm using correlation coefficient. Moreover, to improve the accuracy of co-registration, the outliers in the initial matching point should be removed. For this, two outlier detection techniques of studentized residual and modified RANSAC algorithm are used in this study. Three pairs of Landsat images were used for performance test, and the results were compared and evaluated in terms of robustness and efficiency.

• The Journal of Korean Society for Radiation Therapy
• /
• v.29 no.2
• /
• pp.119-128
• /
• 2017

Purpose: The purpose of this study was to measure the average actual treatment time at the time of Tomotherapy treatment. We want to investigate the time required for the procedure in the treatment process that affects the actual treatment time. Patients and Methods: We measured the time required by the procedure in 31 patients who were treated with tomography therapy. Beam-on time, Image registration time, Set-up with scan time and Actual treatment time were measured and stepwise regression analysis was performed. Result: The average treatment time per a patient was 21.44 - 23.92 minutes. Beam-on time, Image registration time, and Set-up with Scan time were the important factors affecting the actual treatment time. The biggest influence was Beam-on time and Registration time was less affected by analysing. Conclusion: The average treatment time per a patient in tomotherapy treatment was $22.68{\pm}3.37$ minutes. The Approximately 21 patients are expected to be treated within 8 hours of regular work time. However, if the treatment is interrupted or the time of the procedure is changed during the treatment process, it affects the schedule of the daily treatment patients and the workload is expected to increase.

• Korean Journal of Remote Sensing
• /
• v.27 no.6
• /
• pp.693-701
• /
• 2011

Image registration is an essential process to analyze the time series of satellite images for the purpose of image fusion and change detection. The Mutual Information (MI) is commonly used as similarity measure for image registration because of its robustness to noise. Due to the radiometric differences, it is not easy to apply MI to multi-temporal satellite images using directly the pixel intensity. Image features for MI are more abundantly obtained by employing a Gabor filter which varies adaptively with the filter characteristics such as filter size, frequency and orientation for each pixel. In this paper we employed Bidirectional Gabor Filter Energy (BGFE) defined by Gabor filter features and applied the BGFE to similarity measure calculation as an image feature for MI. The experiment results show that the proposed method is more robust than the conventional MI method combined with intensity or gradient magnitude.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2005.05a
• /
• pp.345-349
• /
• 2005

Massive amount of satellite image processing such asglobal/continental-level analysis and monitoring requires automated and speedy georegistration. There could be two major automated approaches: (1) rigid mathematical modeling using sensor model and ephemeris data; (2) heuristic co-registration approach with respect to existing reference image. In case of ETM+, the accuracy of the first approach is known as RMSE 250m, which is far below requested accuracy level for most of satellite image processing. On the other hands, the second approach is to find identical points between new image and reference image and use heuristic regression model for registration. The latter shows better accuracy but has problems with expensive computation. To improve efficiency of the coregistration approach, the author proposed a pre-qualified matching algorithm which is composed of feature extraction with canny operator and area matching algorithm with correlation coefficient. Throughout the pre-qualification approach, the computation time was significantly improved and make the registration accuracy is improved. A prototype was implemented and tested with the proposed algorithm. The performance test of 14 TM/ETM+ images in the U.S. showed: (1) average RMSE error of the approach was 0.47 dependent upon terrain and features; (2) the number average matching points were over 15,000; (3) the time complexity was 12 min per image with 3.2GHz Intel Pentium 4 and 1G Ram.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2006.04a
• /
• pp.181-185
• /
• 2006

Image co-registration is the process of overlaying two images of the same scene, one of which represents a reference image, while the other is geometrically transformed to the one. In order to improve efficiency and effectiveness of the co-registration approach, the author proposed a pre-qualified area matching algorithm which is composed of feature extraction with canny operator and area matching algorithm with cross correlation coefficient. For refining matching points, outlier detection using studentized residual was used and iteratively removes outliers at the level of three standard deviation. Throughout the pre-qualification and the refining processes, the computation time was significantly improved and the registration accuracy is enhanced. A prototype of the proposed algorithm was implemented and the performance test of 3 Landsat images of Korea showed: (1) average RMSE error of the approach was 0.436 Pixel (2) the average number of matching points was over 38,475 (3) the average processing time was 489 seconds per image with a regular workstation equipped with a 3 GHz Intel Pentium 4 CPU and 1 Gbytes Ram. The proposed approach achieved robustness, full automation, and time efficiency.

• Journal of the Institute of Convergence Signal Processing
• /
• v.16 no.1
• /
• pp.9-14
• /
• 2015

Outlier detection and removal is a crucial step needed for various image processing applications such as image registration. Random Sample Consensus (RANSAC) is known to be the best algorithm so far for the outlier detection and removal. However RANSAC requires a cosiderable computation time. To drastically reduce the computation time while preserving the comparable quality, a outlier detection and removal method based on modified K-means is proposed. The original K-means was conducted first for matching point pairs and then cluster merging and member exclusion step are performed in the modification step. We applied the methods to various images with highly repetitive patterns under several geometric distortions and obtained successful results. We compared the proposed method with RANSAC and showed that the proposed method runs 3~10 times faster than RANSAC.