Scale Invariant Feature Transform Method Research Articles

Illumination invariant feature point matching for high-resolution planetary remote sensing images

Despite its success with regular close-range and remote-sensing images, the scale-invariant feature transform (SIFT) algorithm is essentially not invariant to illumination differences due to the use of gradients for feature description. In planetary remote sensing imagery, which normally lacks sufficient textural information, salient regions are generally triggered by the shadow effects of keypoints, reducing the matching performance of classical SIFT. Based on the observation of dual peaks in a histogram of the dominant orientations of SIFT keypoints, this paper proposes an illumination-invariant SIFT matching method for high-resolution planetary remote sensing images. First, as the peaks in the orientation histogram are generally aligned closely with the sub-solar azimuth angle at the time of image collection, an adaptive suppression Gaussian function is tuned to level the histogram and thereby alleviate the differences in illumination caused by a changing solar angle. Next, the suppression function is incorporated into the original SIFT procedure for obtaining feature descriptors, which are used for initial image matching. Finally, as the distribution of feature descriptors changes after anisotropic suppression, and the ratio check used for matching and outlier removal in classical SIFT may produce inferior results, this paper proposes an improved matching procedure based on cross-checking and template image matching. The experimental results for several high-resolution remote sensing images from both the Moon and Mars, with illumination differences of 20°–180°, reveal that the proposed method retrieves about 40%–60% more matches than the classical SIFT method. The proposed method is of significance for matching or co-registration of planetary remote sensing images for their synergistic use in various applications. It also has the potential to be useful for flyby and rover images by integrating with the affine invariant feature detectors.

Palm Pattern Recognition using Scale Invariant Feature Transform (SIFT)

In this research paper, we propose an efficient scale invariant feature transform (SIFT) for palm pattern recognition. A fingerprint recognition which is efficient for individual authentication based on fingerprint pattern. This method leads to fraudulent because it could be extracted easily from individuals. The SIFT method based on feature detection overcomes the above problem and is a combination of fast key point detector and visual descriptor. Using SIFT method contactless palm pattern images can be acquired, matched, recognised, authenticated and their matching performance are simulated using OpenCV. The experimental results show that SIFT method provides significantly fast and improved performance than the conventional methods like oriented FAST and rotated BRIEF (ORB).

Optimization of SIFT algorithm for fast-image feature extraction in line-scanning ophthalmoscope

The Scale Invariant Feature Transform (SIFT) algorithm is utilized broadly in image registration to improve image qualities. However, the algorithm’s complexity reduces its efficiency in biology study and usually requires real-time. In this article, we present an improved SIFT technique in software architecture for matching sequences of images taken from a line-scanning ophthalmoscope (LSO). The method generates the Gaussian Scale-space pyramid in frequency domain to complete the SIFT feature detector more quickly. A novel SIFT descriptor invariable with rotation and illumination is then created to reduce calculation time, implementing the original SIFT method, our improved SIFT method, and the graphic processing unit (GPU) version of our improved SIFT method. The experiments have shown that the improved SIFT is almost 2–3 times faster than the original while maintaining more robust performance, and the GPU implementation of the improved SIFT is 20 times faster than central processing unit (CPU) implementation and achieves acceleration at real-time as expected. Although tested on an LSO system, the improved SIFT method does not rely on the acquisition setup. As a result, this method can be applied to other imaging instruments, e.g., adaptive optics to increase their resolution in agreement.

MBR-SIFT: A mirror reflected invariant feature descriptor using a binary representation for image matching

The traditional scale invariant feature transform (SIFT) method can extract distinctive features for image matching. However, it is extremely time-consuming in SIFT matching because of the use of the Euclidean distance measure. Recently, many binary SIFT (BSIFT) methods have been developed to improve matching efficiency; however, none of them is invariant to mirror reflection. To address these problems, in this paper, we present a horizontal or vertical mirror reflection invariant binary descriptor named MBR-SIFT, in addition to a novel image matching approach. First, 16 cells in the local region around the SIFT keypoint are reorganized, and then the 128-dimensional vector of the SIFT descriptor is transformed into a reconstructed vector according to eight directions. Finally, the MBR-SIFT descriptor is obtained after binarization and reverse coding. To improve the matching speed and accuracy, a fast matching algorithm that includes a coarse-to-fine two-step matching strategy in addition to two similarity measures for the MBR-SIFT descriptor are proposed. Experimental results on the UKBench dataset show that the proposed method not only solves the problem of mirror reflection, but also ensures desirable matching accuracy and speed.

Rat Brain Registration Using Improved Speeded Up Robust Features

This study proposes an approach that combines edge feature extraction (EFE) with speeded up robust features (SURF) to register rat brain slices. When extracting feature points, using the traditional scale-invariant feature transform (SIFT) approach has several shortcomings, including blurry edge information and the inability to accurately extract smooth edge targets. Therefore, a feature point description method that comprises EFE and SURF is proposed to match local feature points, thereby effectively capturing edge feature points and avoiding mistaking grains and noise in images for feature points. In addition, a wavelet feature description approach is used to resolve incorrect matching caused by noise. Feature point matching for local regions is devised to solve mismatch errors that occur in highly similar regions in an image. Compared to a state-of-the-art SIFT method, the proposed method generates better results for rat brain image registration in terms of subjective visual presentation and registration quality criteria.

Experimental analysis of SIFT and SURF features for multi-object image retrieval

Content-based image retrieval (CBIR) is the most challenging task of retrieving similar images for large databases. In this work, we present experimental analysis based on scale invariant feature transform (SIFT) and speeded up robust features (SURF) as local features for multi-object image retrieval. The experiments are conducted on database consisting of group of images obtained by aggregating two objects, three objects, four objects and five objects together from the Columbia Object Image Library (COIL-100) image dataset of ten categories. For the query image, features are extracted using SIFT or SURF. These features of query image are compared with corresponding features stored in the database for similarity retrieval. The precision, recall and FMeasure values are computed for each group and results are tabulated. The experimental results show that the accuracy of retrieval from SURF is better than SIFT method.

Experimental analysis of SIFT and SURF features for multi-object image retrieval

Content-based image retrieval (CBIR) is the most challenging task of retrieving similar images for large databases. In this work, we present experimental analysis based on scale invariant feature transform (SIFT) and speeded up robust features (SURF) as local features for multi-object image retrieval. The experiments are conducted on database consisting of group of images obtained by aggregating two objects, three objects, four objects and five objects together from the Columbia Object Image Library (COIL-100) image dataset of ten categories. For the query image, features are extracted using SIFT or SURF. These features of query image are compared with corresponding features stored in the database for similarity retrieval. The precision, recall and FMeasure values are computed for each group and results are tabulated. The experimental results show that the accuracy of retrieval from SURF is better than SIFT method.

A new modified panoramic UAV image stitching model based on the GA-SIFT and adaptive threshold method

This paper presents panoramic unmanned aerial vehicle (UAV) image stitching techniques based on an optimal Scale Invariant Feature Transform (SIFT) method. The image stitching representation associates a transformation matrix with each input image. In this study, we formulate stitching as a multi-image matching problem, and use invariant local features to find matches between the images. An improved Geometric Algebra (GA-SIFT) algorithm is proposed to realize fast feature extraction and feature matching work for the scanned images. The proposed GA-SIFT method can locate more feature points with greater accurately than the traditional SIFT method. The adaptive threshold value method proposed solves the limitation problem of high computation load and high cost of stitching time by greater feature points extraction and stitching work. The modified random sample consensus method is proposed to estimate the image transformation parameters and to determine the solution with the best consensus for the data. The experimental results demonstrate that the proposed image stitching method greatly increases the speed of the image alignment process and produces a satisfactory image stitching result. The proposed image stitching model for aerial images has good distinctiveness and robustness, and can save considerable time for large UAV image stitching.

Orthogonal design for scale invariant feature transform optimization

To improve object recognition capabilities in applications, we used orthogonal design (OD) to choose a group of optimal parameters in the parameter space of scale invariant feature transform (SIFT). In the case of global optimization (GOP) and local optimization (LOP) objectives, our aim is to show the operation of OD on the SIFT method. The GOP aims to increase the number of correctly detected true matches (NoCDTM) and the ratio of NoCDTM to all matches. In contrast, the LOP mainly aims to increase the performance of recall–precision. In detail, we first abstracted the SIFT method to a 9-way fixed-effect model with an interaction. Second, we designed a mixed orthogonal array, MA(64,23420,2), and its header table to optimize the SIFT parameters. Finally, two groups of parameters were obtained for GOP and LOP after orthogonal experiments and statistical analyses were implemented. Our experiments on four groups of data demonstrate that compared with the state-of-the-art methods, GOP can access more correct matches and is more effective against object recognition. In addition, LOP is favorable in terms of the recall–precision.

A Technical Analysis of Image StitchingAlgorithm Using Different Corner DetectionMethods

An image stitching is a method of combining multiple overlapping images of the same scene into a larger image without loss of information. Literature shows the use of various corner detection algorithms in image stitching. The most widely used are Harris corner detection method and SIFTS (Scale Invariant Feature Transform) method. In this paper, a comparative study is done forHarris corner detection algorithm and SIFT algorithm in image stitching using similarity mat rix matching scheme. Total 30 pairs of different images have been used for simulation and comparison. The algorithms have been compared with respect to number of corner detected, number of matching pairs and matching time. From the simulation results it has been observed that SIFT corner detection method is more efficient in image stitching.

New Analysis Method Application in Metallographic Images through the Construction of Mosaics Via Speeded Up Robust Features and Scale Invariant Feature Transform

In many applications in metallography and analysis, many regions need to be considered and not only the current region. In cases where there are analyses with multiple images, the specialist should also evaluate neighboring areas. For example, in metallurgy, welding technology is derived from conventional testing and metallographic analysis. In welding, these tests allow us to know the features of the metal, especially in the Heat-Affected Zone (HAZ); the region most likely for natural metallurgical problems to occur in welding. The expanse of the Heat-Affected Zone exceeds the size of the area observed through a microscope and typically requires multiple images to be mounted on a larger picture surface to allow for the study of the entire heat affected zone. This image stitching process is performed manually and is subject to all the inherent flaws of the human being due to results of fatigue and distraction. The analyzing of grain growth is also necessary in the examination of multiple regions, although not necessarily neighboring regions, but this analysis would be a useful tool to aid a specialist. In areas such as microscopic metallography, which study metallurgical products with the aid of a microscope, the assembly of mosaics is done manually, which consumes a lot of time and is also subject to failures due to human limitations. The mosaic technique is used in the construct of environment or scenes with corresponding characteristics between themselves. Through several small images, and with corresponding characteristics between themselves, a new model is generated in a larger size. This article proposes the use of Digital Image Processing for the automatization of the construction of these mosaics in metallographic images. The use of this proposed method is meant to significantly reduce the time required to build the mosaic and reduce the possibility of failures in assembling the final image; therefore increasing efficiency in obtaining results and expediting the decision making process. Two different methods are proposed: One using the transformed Scale Invariant Feature Transform (SIFT), and the second using features extractor Speeded Up Robust Features (SURF). Although slower, the SIFT method is more stable and has a better performance than the SURF method and can be applied to real applications. The best results were obtained using SIFT with Peak Signal-to-Noise Ratio = 61.38, Mean squared error = 0.048 and mean-structural-similarity = 0.999, and processing time of 4.91 seconds for mosaic building. The methodology proposed shows be more promissory in aiding specialists during analysis of metallographic images.

A Novel Approach for Object Detection in Illuminated and Occluded Video Sequences Using Visual Information with Object Feature Estimation

This paper reports a novel object-detection technique in video sequences. The proposed algorithm consists of detection of objects in illuminated and occluded videos by using object features and a neural network technique. It consists of two functional modules: region-based object feature extraction and continuous detection of objects in video sequences with region features. This scheme is proposed as an enhancement of the Lowe’s scale-invariant feature transform (SIFT) object detection method. This technique solved the high computation time problem of feature generation in the SIFT method. The improvement is achieved by region-based feature classification in the objects to be detected; optimal neural network?based feature reduction is presented in order to reduce the object region feature dataset with winner pixel estimation between the video frames of the video sequence. Simulation results show that the proposed scheme achieves better overall performance than other object detection techniques, and region-based feature detection is faster in comparison to other recent techniques.

해안 광학영상 자료를 이용한 쇄파지역 연안류 측정기술

In-situ measurements are labor-intensive, time-consuming, and limited in their ability to observe currents with spatial variations in the surf zone. This paper proposes an optical image-based method of measurement of currents in the surf zone. This method measures nearshore currents by tracking in time wave breaking-induced foam patches from sequential images. Foam patches in images tend to be arrayed with irregular pixel intensity values, which are likely to remain consistent for a short period of time. This irregular intensity feature of a foam patch is characterized and represented as a keypoint using an imagebased object recognition method, i.e., Scale Invariant Feature Transform (SIFT). The keypoints identified by the SIFT method are traced from time sequential images to produce instantaneous velocity fields. In order to remove erroneous velocities, the instantaneous velocity fields are filtered by binding them within upper and lower limits, and averaging the velocity data in time and space with a certain interval. The measurements that are obtained by this method are comparable to the results estimated by an existing image-based method of observing currents, named the Optical Current Meter (OCM).

Comparative Study of Corner and Feature Extractors for Real-Time Object Recognition in Image Processing

Corner detection and feature extraction are essential aspects of computer vision problems such as object recognition and tracking. Feature detectors such as Scale Invariant Feature Transform (SIFT) yields high quality features but computationally intensive for use in real-time applications. The Features from Accelerated Segment Test (FAST) detector provides faster feature computation by extracting only corner information in recognising an object. In this paper we have analyzed the efficient object detection algorithms with respect to efficiency, quality and robustness by comparing characteristics of image detectors for corner detector and feature extractors. The simulated result shows that compared to conventional SIFT algorithm, the object recognition system based on the FAST corner detector yields increased speed and low performance degradation. The average time to find keypoints in SIFT method is about 0.116 seconds for extracting 2169 keypoints. Similarly the average time to find corner points was 0.651 seconds for detecting 1714 keypoints in FAST methods at threshold 30. Thus the FAST method detects corner points faster with better quality images for object recognition.

적외선 영상에서 변위추정 및 SURF 특징을 이용한 표적 탐지 분류 기법

본 논문에서는 적외선 영상에서 영상 변위를 이용하여 기동 표적 영역을 탐지하고, SURF(Speeded Up Robust Features) 특징점에 대한 BAS(Beam Angle Statistics)를 이용하여 분류하는 시스템에 대하여 설명한다. 영상 기반 기술 분야에서 대표적인 대응점 정합 알고리즘인 SURF 기법은 SIFT(Scale Invariant Feature Transform) 기법에 비해 정합 속도가 매우 빠르고 비슷한 정합 성능을 보이기 때문에 널리 사용되고 있다. SURF를 이용한 대부분의 객체 인식의 경우 특징점 추출과 정합의 과정을 수행하지만, 제안하는 기법은 표적의 기동 특성을 반영하여 영상의 변위 추정을 통하여 표적의 영역을 탐지하고 SURF 특징점 들의 기하구조를 판단함으로써 표적 분류를 수행한다. 제안하는 기법은 무인 표적 탐지/인지 시스템의 초기모델 구축을 위하여 연구가 진행되었으며, 모의 표적을 이용한 가상 영상과 적외선 실 영상을 이용하여 실험한 결과 약 73~85%의 분류 성능을 확인하였다. In this paper, we propose the target detection method using image displacement, and classification method using SURF(Speeded Up Robust Features) feature points and BAS(Beam Angle Statistics) in infrared images. The SURF method that is a typical correspondence matching method in the area of image processing has been widely used, because it is significantly faster than the SIFT(Scale Invariant Feature Transform) method, and produces a similar performance. In addition, in most SURF based object recognition method, it consists of feature point extraction and matching process. In proposed method, it detects the target area using the displacement, and target classification is performed by using the geometry of SURF feature points. The proposed method was applied to the unmanned target detection/recognition system. The experimental results in virtual images and real images, we have approximately 73~85% of the classification performance.

Parameter Optimization for the Extraction of Matching Points Between High-Resolution Multisensor Images in Urban Areas

The objective of this paper is to extract a suitable number of evenly distributed matched points, given the characteristics of the site and the sensors involved. The intent is to increase the accuracy of automatic image-to-image registration for high-resolution multisensor data. The initial set of matching points is extracted using a scale-invariant feature transform (SIFT)-based method, which is further used to evaluate the initial geometric relationship between the features of the reference and sensed images. The precise matching points are extracted considering location differences and local properties of features. The values of the parameters used in the precise matching are optimized using an objective function that considers both the distribution of the matching points and the reliability of the transformation model. In case studies, the proposed algorithm extracts an appropriate number of well-distributed matching points and achieves a higher correct-match rate than the SIFT method. The registration results for all sensors are acceptably accurate, with a root-mean-square error of less than 1.5 m.

Registration of Building Intensity Image and its Color Image Using SIFT Algorithm

Fusion of close range photogrammetry (CRP) and terrestrial laser scanning (TLS) technology has been a hot topic in the field of building reconstruction. There are many ways to realize the fusion of the two kind data. In this paper, we propose a method for 3D-2D data registration based on Scale Invariant Feature Transform (SIFT) algorithm and range intensity data. 3D terrestrial laser scanner and digital camera are different sensors, which will lead to large difference between intensity image (derived from range intensity data) and color image. The traditional image matching method can not apply to register these kind images. This paper focuses on studying the feasibility and practicability of SIFT algorithm on such different images matching. The result shows that the principal of SIFT method is suitable for the registration of the two kind images.

Implementation of Image Matching Algorithm Based on SIFT Features

With the aim to solve the implement problem in scale invariant feature transform (SIFT) algorithm, the theory and the implementation process was analyzed in detail. The characteristics of the SIFT method were analyzed by theory, combined with the explanation of the Rob Hess SIFT source codes. The effect of the SIFT method was validated by matching two different real images. The matching result shows that the features extracted by SIFT method have excellent adaptive and accurate characteristics to image scale, viewpoint change, which are useful for the fields of image recognition, image reconstruction, etc.

SU-E-J-237: Image Feature Based DRR and Portal Image Registration

Purpose: Two-dimensional (2D) matching of the kV X-ray and digitally reconstructed radiography (DRR) images is an important setup technique for image-guided radiotherapy (IGRT). In our clinics, mutual information based methods are used for this purpose on commercial linear accelerators, but with often needs for manual corrections. This work proved the feasibility that feature based image transform can be used to register kV and DRR images. Methods: The scale invariant feature transform (SIFT) method was implemented to detect the matching image details (or key points) between the kV and DRR images. These key points represent high image intensity gradients, and thus the scale invariant features. Due to the poor image contrast from our kV image, direct application of the SIFT method yielded many detection errors. To assist the finding of key points, the center coordinates of the kV and DRR images were read from the DICOM header, and the two groups of key points with similar relative positions to their corresponding centers were paired up. Using these points, a rigid transform (with scaling, horizontal and vertical shifts) was estimated. We also artificially introduced vertical and horizontal shifts to test the accuracy of our registration method on anterior-posterior (AP) and lateral pelvic images. Results: The results provided a satisfactory overlay of the transformed kV onto the DRR image. The introduced vs. detected shifts were fit into a linear regression. In the AP image experiments, linear regression analysis showed a slope of 1.15 and 0.98 with an R2 of 0.89 and 0.99 for the horizontal and vertical shifts, respectively. The results are 1.2 and 1.3 with R2 of 0.72 and 0.82 for the lateral image shifts. Conclusion: This work provided an alternative technique for kV to DRR alignment. Further improvements in the estimation accuracy and image contrast tolerance are underway.

SIFT algorithm-based 3D pose estimation of femur

To address the lack of 3D space information in the digital radiography of a patient femur, a pose estimation method based on 2D-3D rigid registration is proposed in this study. The method uses two digital radiography images to realize the preoperative 3D visualization of a fractured femur. Compared with the pure Digital Radiography or Computed Tomography imaging diagnostic methods, the proposed method has the advantages of low cost, high precision, and minimal harmful radiation. First, stable matching point pairs in the frontal and lateral images of the patient femur and the universal femur are obtained by using the Scale Invariant Feature Transform method. Then, the 3D pose estimation registration parameters of the femur are calculated by using the Iterative Closest Point (ICP) algorithm. Finally, based on the deviation between the six degrees freedom parameter calculated by the proposed method, preset posture parameters are calculated to evaluate registration accuracy. After registration, the rotation error is less than l.5°, and the translation error is less than 1.2 mm, which indicate that the proposed method has high precision and robustness. The proposed method provides 3D image information for effective preoperative orthopedic diagnosis and surgery planning.