Scale Invariant Feature Transform Feature Matching Research Articles

Artificial intelligence scale-invariant feature transform algorithm-based system to improve the calculation accuracy of Ki-67 index in invasive breast cancer: a multicenter retrospective study.

Ki-67 is a key indicator of the proliferation activity of tumors. However, no standardized criterion has been established for Ki-67 index calculation. Scale-invariant feature transform (SIFT) algorithm can identify the robust invariant features to rotation, translation, scaling and linear intensity changes for matching and registration in computer vision. Thus, this study aimed to develop a SIFT-based computer-aided system for Ki-67 calculation in breast cancer. Hematoxylin and eosin (HE)-stained and Ki-67-stained slides were scanned and whole slide images (WSIs) were obtained. The regions of breast cancer (BC) tissues and non-BC tissues were labeled by experienced pathologists. All the labeled WSIs were randomly divided into the training set, verification set, and test set according to a fixed ratio of 7:2:1. The algorithm for identification of cancerous regions was developed by a ResNet network. The registration process between paired consecutive HE-stained WSIs and Ki-67-stained WSIs was based on a pyramid model using the feature matching method of SIFT. After registration, we counted the nuclear-stained Ki-67-positive cells in each identified invasive cancerous region using color deconvolution. To assess the accuracy, the AI-assisted result for each slice was compared with the manual diagnosis result of pathologists. If the difference of the two positive rate values is not greater than 10%, it was a consistent result; otherwise, it was an inconsistent result. The accuracy of the AI-based algorithm in identifying breast cancer tissues in HE-stained slides was 93%, with an area under the curve (AUC) of 0.98. After registration, we succeeded in identifying Ki-67-positive cells among cancerous cells across the entire WSIs and calculated the Ki-67 index, with an accuracy rate of 91.5%, compared to the gold standard pathological reports. Using this system, it took about 1 hour to complete the evaluation of all the tested 771 pairs of HE- and Ki-67-stained slides. Each Ki-67 result took less than 2 seconds. Using a pyramid model and the SIFT feature matching method, we developed an AI-based automatic cancer identification and Ki-67 index calculation system, which could improve the accuracy of Ki-67 index calculation and make the data repeatable among different hospitals and centers.

An improved smart key frame extraction algorithm for vehicle target recognition

The smart extraction of key frames is one of the critical technologies for vehicle target recognition in the video streaming environment. Currently, the common algorithms are scale-invariant feature transform (SIFT) or background-difference method. Both algorithms have their own advantages and shortcomings. In this paper, a smart key frame extraction algorithm is proposed by combining the background difference method and SIFT feature matching algorithm, at the same time, the criterion factor K is introduced. The algorithm absorbs the idea of background difference and SIFT feature matching, and divides video frames into mutation frames and non-mutation frames. Hash algorithm is used to distinguish the similarity of two frames and remove redundant frames. The local feature of video sequence is introduced to solve the problem of sharp increasing calculation for inter frame distance. Experimental results show that the improved smart key frame extraction algorithm has a lower missing frame rate and higher accuracy for vehicle target recognition.

Adaptive depth constraint-based underwater binocular image feature matching

Objective In this paper, addressing sparse feature points and unique epipolar constraint, an adaptive depth constraint-based underwater feature matching (ADC-UFM) scheme is proposed. Methods By combining the features from accelerated segment test (FAST) operator with scale invariant feature transform (SIFT) descriptors such that the matching accuracy can be dramatically improved. By introducing the underwater refractive factor, the matching constraint model (MCM) can be effectively established, and thereby contributing to eliminating mismatched points. The adaptive threshold choosing (ATC) module is finely devised to extremely preserve image feature information in changeable underwater environments. Results Comprehensive experiments show that the proposed ADC-UFM scheme can outperform typical matching schemes including SIFT, speeded-up robust features (SURF) and SIFT feature matching based on underwater curve constraint (UCC-SIFT), which not only achieves 85.2% matching accuracy but also meets real-time requirements. Conclusion The results of this study can provide a reliable guarantee for the subsequent underwater 3D reconstruction based on binocular vision system.

Integrating SIFT and CNN Feature Matching for Partial-Duplicate Image Detection

With the increasing popularity of various deep neural networks in the area of computational intelligence, the research attention for content-based image detection/retrieval has been shifted from the handcrafted local features such as scale invariant feature transform (SIFT) to the features derived from convolutional neural networks (CNN). However, the existing image-based CNN features, directly extracted from the entire images, are not suitable for detecting small duplicate regions, while region-based CNN features show limited robustness to a variety of image modifications such as rescaling, occlusion, and noise adding. These will affect the performance of partial-duplicate image detection. To address these issues, we propose an integrated feature matching scheme, which integrates the matching of SIFT features and CNN features between images for partial-duplicate image detection. In this scheme, we first implement SIFT feature matching based on the bag-of-visual-words model to detect the potential duplicate region pairs between images, and then match the CNN features of these regions extracted from the deep convolutional layer of CNN to compute image similarity. Since both the good robustness of SIFT features and the high discriminative power of CNN features are sufficiently explored, our scheme allows an accurate detection. Experimental results show that the proposed approach provides superior accuracy than the state of the arts, while achieves comparable efficiency.

Remote Sensing Image Registration Based on Modified SIFT and Feature Slope Grouping

In feature-based remote sensing image registration, the scale-invariant feature transform (SIFT) algorithm has been one of the most popular solutions. However, it is still a challenge to possess an appropriate amount of correct matches while eliminating mismatches. In this letter, inspired by SIFT, an accurate and robust feature matching framework based on feature slope grouping (FSG) for remote sensing image registration is proposed. Our FSG-SIFT algorithm consists of four major phases: modified SIFT, feature slope computation, feature point grouping, and outlier removal and transformation. Specifically, the random sample consensus is adopted to refine the matches followed by the affine transform. The proposed remote sensing image registration algorithm has been validated on a wide variety of high-resolution orthoimagery data. Experimental results with multispectral and multitemporal images suggested that this new image registration algorithm well improved the feature matching accuracy with better registration performance over five state-of-the-art methods.

Efficient Biometric Recognition Methodology using Guided Filtering and SIFT Feature Matching

A novel infrared finger vein biometric identification is proposed using Linear Gabor filter with Guidance image and SIFT feature matching. Linear Gabor filter with guidance image is used for extracting finger vein pattern without segmentation processing and also performs well with some poor quality images due to low contrast, illuminance imbalance or noise etc. Firstly, we utilized Guided Linear Gabor filter for ridge detection as simple Linear Gabor filter and also enhance the image by performing edge preserving smoothing operation. Secondly we utilized SIFT feature matching for verification. A SIFT (Scale Invariant Feature Transform) can extract features to posses rotation invariance and shift invariance for providing better matching rate. The simulation analysis shows our proposed system is an effective feature for finger vein biometric identification.

Automatic adaptation of SIFT for robust facial recognition in uncontrolled lighting conditions

The scale invariant feature transform (SIFT), which was proposed by David Lowe, is a powerful method that extracts and describes local features called keypoints from images. These keypoints are invariant to scale, translation, and rotation, and partially invariant to image illumination variation. Despite their robustness against these variations, strong lighting variation is a difficult challenge for SIFT‐based facial recognition systems, where significant degradation of performance has been reported. To develop a robust system under these conditions, variation in lighting must be first eliminated. Additionally, SIFT parameter default values that remove unstable keypoints and inadequately matched keypoints are not well‐suited to images with illumination variation. SIFT keypoints can also be incorrectly matched when using the original SIFT matching method. To overcome this issue, the authors propose propose a method for removing the illumination variation in images and correctly setting SIFT's main parameter values (contrast threshold, curvature threshold, and match threshold) to enhance SIFT feature extraction and matching. The proposed method is based on an estimation of comparative image lighting quality, which is evaluated through an automatic estimation of gamma correction value. Through facial recognition experiments, the authors find significant results that clearly illustrate the importance of the proposed robust recognition system.

Multiple Vehicle Detection and Tracking in Highway Traffic Surveillance Video Based on SIFT Feature Matching

This paper presents a complete method for vehicle detection and tracking in a fixed setting based on computer vision. Vehicle detection is performed based on Scale Invariant Feature Transform (SIFT) feature matching. With SIFT feature detection and matching, the geometrical relations between the two images is estimated. Then, the previous image is aligned with the current image so that moving vehicles can be detected by analyzing the difference image of the two aligned images. Vehicle tracking is also performed based on SIFT feature matching. For the decreasing of time consumption and maintaining higher tracking accuracy, the detected candidate vehicle in the current image is matched with the vehicle sample in the tracking sample set, which contains all of the detected vehicles in previous images. Most remarkably, the management of vehicle entries and exits is realized based on SIFT feature matching with an efficient update mechanism of the tracking sample set. This entire method is proposed for highway traffic environment where there are no non-automotive vehicles or pedestrians, as these would interfere with the results.

Mean Shift tracking for video moving objects in combination with scale invariant feature transform and Kalman filter

To solve the problem of poor tracking performance when the moving target has a relatively large scale change,rotation, fast-moving or occlusion, an object tracking method combining Scale Invariant Feature Transform( SIFT) matching and Kalman filter with the Mean Shift algorithm was put forward. First, the Kalman filter was used to predict the movement state of the moving target and its estimated value was taken as the initial position of Mean Shift tracking. Then, when the measure coefficient for the similarity of the candidate target model and the initial target model was less than a certain threshold, SIFT feature matching was used to look for the possible position of the target and the new candidate target model was built there, meanwhile, the similarity with the initial target model was measured. Finally, by comparing the two matching coefficients, the position associated with a larger one was selected as the target's final position. The experimental results show that the average tracking error of this algorithm is decreased by about twenty percent than the tracking algorithms only combining the SIFT feature or Kalman filter with the Mean Shift alone.

The Study of Reduce the Mismatch Based on the SIFT Feature Matching

The algorithm of SIFT (scale-invariant feature transform) feature matching is an international hotspot in the areas of the keypoints matching and target recognition at the present time. The algorithm is used in the image matching widely because of the good invariance of scale, illumination and space rotation .This paper proposes a new theory to reduce the mismatch—the theory to reduce the mismatch based on the main orientation of keypoints. This theory should firstly compute the grads of the main orientation of a couple of matched keypoints in the two images and the difference between them. Because the difference of the main orientation of matched keypoints should be much larger than the couples which are matched correctly, we can distinguish and reduce the mismatch through setting the proper threshold, and it can improve the accuracy of the SIFT algorithm greatly.

The Application of SIFT Algorithm in Blind Road Environmental Image Matching

Image Matching is a key technology in the intelligent navigation system for the blind, which is based on the computer video. The images of moving blind people, collected at real time, have variety of changes in light, rotation, scaling, etc. Against this feature, we propose a practical matching algorithm, which is based on the SIFT (Scale Invariant, Feature Transform). That is the image matching algorithm. We focus on the algorithm of the SIFT feature extraction and matching, and obtain the feature points of the image through feature extraction algorithm. We verify the effect of the algorithm by selecting practical images with rotation, scaling and different light. The result is that this method can get better matches for the blind road environmental image.

Monocular Vision Based Mobile Robot 3D Map Building

A robust dense 3D feature map is built only with monocular vision and odometry. Monocular vision mounted on the robot front-end tracks the 3D natural landmarks, which are structured with matching Scale Invariant Feature Transform (SIFT) feature matching pairs. SIFT features are highly distinctive and invariant to image scaling, rotation, and change in 3D viewpoints. A fast SIFT feature matching algorithm is implemented with the KD-Tree based nearest search approach in the time cost of O(log2N), and matches with large error are eliminated by epipolar line restriction. A map building algorithm based on 3D spatial SIFT landmarks is designed and implemented. Experiment results on Pioneer mobile robot in a real indoor environment show the superior performance of our proposed method.

On the Suitability of SIFT Technique to Deal with Image Modifications Specific to Confocal Scanning Laser Microscopy

Computer vision tasks such as recognition and classification of objects and structures or image registration and retrieval can provide significant information when applied to microscopy images. Recently developed techniques for the detection and description of local features make the extraction and description of local image features that are invariant to various changes possible. The invariance and robustness of feature detection and description techniques play a key role in the design and implementation of object recognition, image registration, or image mosaicing applications. The scale-invariant feature transform (SIFT) technique is a widely used method for the detection, description, and matching of image features. In this article we present the results of our experiments regarding the repeatability of SIFT features, and to the precision of the SIFT feature matching, under image modifications specific to confocal scanning laser microscopy (CSLM). We have analyzed the behavior of SIFT while changing the pinhole aperture, photomultiplier gain, laser beam power, and electronic zoom. Our experiments, conducted on CSLM images, show that the SIFT technique is able to match detected key points between images acquired at different values of the acquisition parameters with good precision and represents a consistent tool for computer vision applications in CSLM.

Automatic mosaic of surveillance images based on SIFT feature matching

In this paper, an improved Scale Invariant Feature Transform (SIFT) feature-based approach for efficient automatic mosaic surveillance images was proposed. In the image registration step, SIFT Feature was extracted efficiently and reliable matching was carried out, and Random Sample Consensus (RANSAC) was improved to guarantee stability and decide the transformation parameters of the images stitching. In the image fuse step, a radiometric and color compensation stage allows the development of an automatic and seamless mosaicing system. The results show that our approach performs well, even in processing the surveillance images with low overlapping area, moving objects, shape differences and noise.