Local Feature Points Research Articles

Unconstrained Human face Tracking in live Video

In surveillance applications visual face detection and tracking becomes an essential task. Many algorithms and technologies have been developed to automatically monitor pedestrians or other moving objects and to track the detected face. One main difficulty in face tracking, among many others, is to choose suitable features and models for detecting and tracking the target. For tracking of faces there are some common features are considered like color, intensity, shape and feature points. In this paper we discuss about mean shift based face tracking based on the color, optical flow tracking based on the intensity and motion, SIFT face tracking based on scale invariant local feature points. Mean shift is then combined with local feature points. Initial results from tries have shown that the implemented method is able to track target face with different pose variation, rotation, partial occlusion and deformation.

Depth recovery for unstructured farmland road image using an improved SIFT algorithm

Road visual navigation relies on accurate road models. This study was aimed at proposing an improved scale-invariant feature transform (SIFT) algorithm for recovering depth information from farmland road images, which would provide a reliable path for visual navigation. The mean image of pixel value in five channels (R, G, B, S and V) were treated as the inspected image and the feature points of the inspected image were extracted by the Canny algorithm, for achieving precise location of the feature points and ensuring the uniformity and density of the feature points. The mean value of the pixels in 5×5 neighborhood around the feature point at an interval of 45o in eight directions was then treated as the feature vector, and the differences of the feature vectors were calculated for preliminary matching of the left and right image feature points. In order to achieve the depth information of farmland road images, the energy method of feature points was used for eliminating the mismatched points. Experiments with a binocular stereo vision system were conducted and the results showed that the matching accuracy and time consuming for depth recovery when using the improved SIFT algorithm were 96.48% and 5.6 s, respectively, with the accuracy for depth recovery of –7.17%-2.97% in a certain sight distance. The mean uniformity, time consuming and matching accuracy for all the 60 images under various climates and road conditions were 50%-70%, 5.0-6.5 s, and higher than 88%, respectively, indicating that performance for achieving the feature points (e.g., uniformity, matching accuracy, and algorithm real-time) of the improved SIFT algorithm were superior to that of conventional SIFT algorithm. This study provides an important reference for navigation technology of agricultural equipment based on machine vision. Keywords: scale-invariant feature transform (sift), feature matching, canny operator, energy method of feature point, farmland road, depth recovery, visual navigation DOI: 10.25165/j.ijabe.20191204.4821 Citation: Yao L J, Hu D, Yang Z D, Li H B, Qian M B. Depth recovery for unstructured farmland road image using an improved SIFT algorithm. Int J Agric & Biol Eng, 2019; 12(4): 141–147.

Deformable registration of x ray and MRI for postimplant dosimetry in low dose rate prostate brachytherapy.

Dosimetric assessment following permanent prostate brachytherapy (PPB) commonly involves seed localization using CT and prostate delineation using coregistered MRI. However, pelvic CT leads to additional imaging dose and requires significant resources to acquire and process both CT and MRI. In this study, we propose an automatic postimplant dosimetry approach that retains MRI for soft-tissue contouring, but eliminates the need for CT and reduces imaging dose while overcoming the inconsistent appearance of seeds on MRI with three projection x rays acquired using a mobile C-arm. Implanted seeds are reconstructed using x rays by solving a combinatorial optimization problem and deformably registered to MRI. Candidate seeds are located in MR images using local hypointensity identification. X ray-based seeds are registered to these candidate seeds in three steps: (a) rigid registration using a stochastic evolutionary optimizer, (b) affine registration using an iterative closest point optimizer, and (c) deformable registration using a local feature point search and nonrigid coherent point drift. The algorithm was evaluated using 20 PPB patients with x rays acquired immediately postimplant and T2-weighted MR images acquired the next day at 1.5T with mean 0.8×0.8×3.0mm voxel dimensions. Target registration error (TRE) was computed based on the distance from algorithm results to manually identified seed locations using coregistered CT acquired the same day as the MRI. Dosimetric accuracy was determined by comparing prostate D90 determined using the algorithm and the ground truth CT-based seed locations. The mean±standard deviation TREs across 20 patients including 1774 seeds were 2.23±0.52mm (rigid), 1.99±0.49mm (rigid+affine), and 1.76±0.43mm (rigid+affine+deformable). The corresponding mean±standard deviation D90 errors were 5.8±4.8%, 3.4±3.4%, and 2.3±1.9%, respectively. The mean computation time of the registration algorithm was 6.1s. The registration algorithm accuracy and computation time are sufficient for clinical PPB postimplant dosimetry.

Maize Leaf Disease Image Classification Using Bag of Features

Image classification is an image grouping based on similarities features. The features extraction stage is a crucial factor for classifying an image. In the conventional image classification, the features commonly used are morphology, color, and texture with various derivative features. The type and number of appropriate features will affect the classification results. In this study, image classification by using the Bag of Features (BOF) method which can generate features automatically. It consists of 4 stages: feature point location by using grid method, feature extraction by using Speed Up Robust Feature (SURF), clustering word-visual vocabularies by using k-means, and classification by using Support Vector Machine (SVM). The classification data are maize leaf images from the PlantVillage-Dataset. The data consists of 3 types of images: RGB, grayscale and segmentation images. Each type includes four classes: healthy, Cercospora, common rust, and northern leaf blight. There are 50 images for each class. We used two scenarios of testing for each type of data: training and validation, 70% and 80% images for training, and the rest for validation. Experimental results showed that the validation accuracies of RGB, grayscale, and segmentation images were 82%, 77%, and 85%.

Underwater Image Enhancement Using Adaptive Retinal Mechanisms.

We propose an underwater image enhancement model inspired by the morphology and function of the teleost fish retina. We aim to solve the problems of underwater image degradation raised by the blurring and nonuniform color biasing. In particular, the feedback from color-sensitive horizontal cells to cones and a red channel compensation are used to correct the nonuniform color bias. The center-surround opponent mechanism of the bipolar cells and the feedback from amacrine cells to interplexiform cells then to horizontal cells serve to enhance the edges and contrasts of the output image. The ganglion cells with color-opponent mechanism are used for color enhancement and color correction. Finally, we adopt a luminance-based fusion strategy to reconstruct the enhanced image from the outputs of ON and OFF pathways of fish retina. Our model utilizes the global statistics (i.e., image contrast) to automatically guide the design of each low-level filter, which realizes the self-adaption of the main parameters. Extensive qualitative and quantitative evaluations on various underwater scenes validate the competitive performance of our technique. Our model also significantly improves the accuracy of transmission map estimation and local feature point matching using the underwater image. Our method is a single image approach that does not require the specialized prior about the underwater condition or scene structure.

Multi-feature analysis fatigue driving based on Conditional Local Neural Fields algorithm detection method

In order to effectively strengthen the monitoring technology of fatigue driving and reduce the incidence of traffic accidents, this paper proposes a new method based on conditional local neural fields algorithm for comprehensive evaluation of fatigue driving state. At present, most of the fatigue driving detection algorithms are based on extracting a single characteristic index, which is strict in environmental requirements and not high in detection. In this paper, the HOG feature is combined with the CLNF algorithm to implement face detection and feature point localization. Then, the EPnP algorithm is used to estimate the head anomaly frequency based on the feature point information, and the blink frequency is calculated according to the EAR eye length aspect ratio concept according to the feature points around the eye. Finally the threshold set by the P80 fatigue detection standard in the PERCLOS method is integrated, and the distributed information fusion strategy is used for fatigue evaluation. Experimental results confirm the effectiveness of the method.

Interpolatory Curve Modeling with Feature Points Control

In curve modeling, interpolation allows users to directly control the location of curve features. While previous literatures have focused on generating interpolatory curves with properties of smoothness and locality, they usually have difficulty in controlling over geometric feature points (e.g., cusps, loops, and inflection points) that mark salient intrinsic features of curve shapes. In this paper, we propose an intuitive and efficient method for constructing planar cubic curves that are curvature continuous almost-everywhere and interpolate a sequence of input data points. Our method provides a good control over the location and type of the geometric feature points. In particular, cusps and loops only occur at specified input data points, while inflection points only occur at specified input data points and joints. We refer to such feature points controlled interpolatory curves as FPC-Curves for short. To construct FPC-Curves, we focus on piecewise cubic curves, where the occurrence of loops, cusps and inflection points is mutually exclusive. We also provide a simple yet efficient algorithm for real-time interactive design of cubic FPC-Curves. Various experimental results show the efficacy and flexibility of our new approach for curve modeling.

Emotion Recognition by a Hybrid System Based on the Features of Distances and the Shapes of the Wrinkles

Abstract Emotion recognition is a key work of research area in brain computer interactions. With the increasing concerns about affective computing, emotion recognition has attracted more and more attention in the past decades. Focusing on geometric positions of key parts of the face and well detecting them is the best way to increase accuracy of emotion recognition systems and reach high classification rates. In this paper, we propose a hybrid system based on wavelet networks using 1D Fast Wavelet Transform. This system combines two approaches: the biometric distances approach where we propose a new technique to locate feature points and the wrinkles approach where we propose a new method to locate the wrinkles regions in the face. The classification rates given by experimental results show the effectiveness of our proposed approach compared to other methods.

Modeling deviation propagation of compliant assembly considering form defects based on basic deviation fields

PurposeA new deviation propagation model considering the form defects in compliant assembly process is proposed. The purpose of this paper is to analyze the deviation propagation by using the basic deviation fields. In particular, each basic deviation field is defined with a generalized compliance matrix of part.Design/methodology/approachFirst, the form defects of parts may be decomposed into a linear combination of basic deviation fields, which are constructed by the eigen-decomposition of the structure stiffness of parts with ideal dimensions. Each basic deviation field is defined with a generalized compliance of part. Moreover, by analyzing the relationship between the basic deviation fields before and after assembling process, a new sensitive matrix is obtained in which each value expresses the correlation of a basic deviation field between the parts and the assembly.FindingsThis model may solve the deviation propagation problems of compliant assembly with considering form defects. Here, one case is used to illustrate the deviation propagation in the assembly process. The results indicate that the proposed method has higher accuracy than the method of influence coefficient when the entire deviation fields of parts are considered. Moreover, the numerical results with the proposed method basically agree with the experimental measurements.Research limitations/implicationsOwing to the hypothesis of linear superposition of basic deviation fields, the research in this paper is limited to the parts with linear elastic deformation. However, the entire form defects of parts are considered rather than the deviations of the local feature points. It may be extended to analyze the three-dimensional deviations of complex thin-walled parts.Originality/valueA deviation propagation model considering parts form defects is developed to achieve more accurate predictions of assembly deviation by using the basic deviation fields.

High-accuracy calibration of line-structured light vision sensor by correction of image deviation.

The structured-light vision sensor calibration approach based on planar target fails to generate reliable and accurate results due to the inaccurate localization of feature points in outdoor complex lighting environments. To address this issue, a novel high-accuracy calibration method that corrects image deviation is proposed for line-structured light vision sensor in this paper. The mathematical solution for stripe point location uncertainty is put forward and the location uncertainty of target feature points and stripe points is established. Moreover, the location deviation of all points is computed through large-scale nonlinear multi-step optimization based on the constraints of uncertainty. After compensation, the planar target based calibration method is adopted to solve the light plane equation. Both simulative and physical experiments have carried out to evaluate the performance of the proposed method and the results show that the proposed method is robust under large feature points localization deviation and can achieve the same measurement accuracy as the planar target calibration method under ideal imaging conditions, which reduces the requirements of the calibration environment, and has important practical engineering application value.

Research on key technologies of intelligent transportation based on image recognition and anti-fatigue driving

Intelligent transportation system needs to solve the main problems in traffic safety. This paper focuses on the traffic safety caused by fatigue driving based on image recognition of key technologies for research and analysis. This paper proposes that the location of face and facial feature points and the classification of fatigue detection are the key links to determine the fatigue driving detection rate. In the analysis of face localization algorithm based on skin color modeling, a corner-based optimization method is proposed to optimize the face region. Based on the analysis of the binary algorithm of human eye localization algorithm, a bi-directional integral projection method is proposed to achieve accurate human eye localization. Then the commonly used fatigue classification algorithm (KNN algorithm) is analyzed. Finally, the proposed method is verified by the simulation test of fatigue driving. Experimental results show that the algorithm based on skin color modeling can accurately locate the driver’s face region. The eye location algorithm based on the two-valued algorithm can also locate the eye location of the tester accurately. The accuracy of KNN fatigue detection model is 87.82%. It can identify driver’s fatigue state with high accuracy.

Framework for automated registration of UAV and UGV point clouds using local features in images

Abstract Automatically registering 3D point clouds generated by unmanned aerial and ground vehicles (UAVs and UGVs) is challenging, as data is acquired at different locations with different sensors, consequently resulting in different spatial scales and occlusions. To address these problems, this study proposes a framework for the automated registration of UAV and UGV point clouds using 2D local feature points in the images taken from UAVs and UGVs. This study first conducted field experiments by varying the angles of the UAV camera to identify the optimal angle with which to detect sufficient points matching with the images taken by the UGV. As a result, this study identified that a combination of UAV images taken at 30° and 90° is appropriate for generating a sufficient number of matching points and attaining a reasonable level of precision. The UAV and UGV point clouds were initially scaled and registered with a transformation matrix computed from the 3D points corresponding to the 2D feature matching points. The initially aligned point clouds were subsequently adjusted by the Iterative Closest Point (ICP) algorithm, resulting in the root mean square error (RMSE) of 0.112 m. This promising result indicates that full automation of spatial data collection and registration from a scattered environment (e.g., construction or disaster sites) by UAVs and UGVs is feasible without human intervention.

A Novel 3D Video Fingerprinting Algorithm Based on Local Feature Points

Digital rights management of the 3D contents is a crucial open issue in the 3D video industry. A novel robust fingerprinting algorithm is proposed for protecting the copyright of the 3D video. Unlike the existing algorithms extracting visual features separately from the 2D videos and the depth maps, in our algorithm a novel local stereo space is constructed according to the depth information of the pixels around the extracted local feature points in the 2D videos, and the 3D videos are processed in a holistic manner. In the proposed space, the 3D-transform-feature is extracted and aggregated into a feature matrix, and then the compact 3D video fingerprints are obtained from the eigenspace of the matrix. Our comprehensive experiments are conducted on a 3D video database, and the results have demonstrated the robustness and discrimination of the proposed algorithm. Moreover, our fingerprints cost less storage spaces than the existing approaches.

LUIFT: LUminance Invariant Feature Transform

Illumination-invariant method for computing local feature points and descriptors, referred to as LUminance Invariant Feature Transform (LUIFT), is proposed. The method helps us to extract the most significant local features in images degraded by nonuniform illumination, geometric distortions, and heavy scene noise. The proposed method utilizes image phase information rather than intensity variations, as most of the state-of-the-art descriptors. Thus, the proposed method is robust to nonuniform illuminations and noise degradations. In this work, we first use the monogenic scale-space framework to compute the local phase, orientation, energy, and phase congruency from the image at different scales. Then, a modified Harris corner detector is applied to compute the feature points of the image using the monogenic signal components. The final descriptor is created from the histograms of oriented gradients of phase congruency. Computer simulation results show that the proposed method yields a superior feature detection and matching performance under illumination change, noise degradation, and slight geometric distortions comparing with that of the state-of-the-art descriptors.

Directional Correlation Filter Bank for Robust Head Pose Estimation and Face Recognition

During the past few decades, face recognition has been an active research area in pattern recognition and computer vision due to its wide range of applications. However, one of the most challenging problems encountered by face recognition is the difficulty of handling large head pose variations. Therefore, the efficient and effective head pose estimation is a critical step of face recognition. In this paper, a novel feature extraction framework, called Directional Correlation Filter Bank (DCFB), is presented for head pose estimation. Specifically, in the proposed framework, the 1-Dimensional Optimal Tradeoff Filters (1D-OTF) corresponding to different head poses are simultaneously and jointly designed in the low-dimensional linear subspace. Different from the traditional methods that heavily rely on the precise localization of the key facial feature points, our proposed framework exploits the frequency domain of the face images, which effectively captures the high-order statistics of faces. As a result, the obtained features are compact and discriminative. Experimental results on public face databases with large head pose variations show the superior performance obtained by the proposed framework on the tasks of both head pose estimation and face recognition.

Ultrasound Environment Classification Based on Fuzzy Model

The fuzzy model based on the centreline ultrasound signal is pro-posed to solve the poor orientation of ultrasonic sensors in the environment modelling in this paper. The locations of ultrasonic arc feature points were extracted and distinguished. For the problem of poor resolution of split type ultrasonic sensors, fuzzy algorithm is applied to identify the obstacle such as line, corner, or arc structure. Ultrasound environment map based on fuzzy is constructed and the maximum distance and angle deviation is 1.2cm and 3°, showing that the accuracy of obstacle classification based on ultrasound and Fuzzy is improved.

Building Detection from VHR Remote Sensing Imagery Based on the Morphological Building Index

Automatic detection of buildings from very high resolution (VHR) satellite images is a current research hotspot in remote sensing and computer vision. However, many irrelevant objects with similar spectral characteristics to buildings will cause a large amount of interference to the detection of buildings, thus making the accurate detection of buildings still a challenging task, especially for images captured in complex environments. Therefore, it is crucial to develop a method that can effectively eliminate these interferences and accurately detect buildings from complex image scenes. To this end, a new building detection method based on the morphological building index (MBI) is proposed in this study. First, the local feature points are detected from the VHR remote sensing imagery and they are optimized by the saliency index proposed in this study. Second, a voting matrix is calculated based on these optimized local feature points to extract built-up areas. Finally, buildings are detected from the extracted built-up areas using the MBI algorithm. Experiments confirm that our proposed method can effectively and accurately detect buildings in VHR remote sensing images captured in complex environments.

Study on a Method of Horizontal Position Measurement for Suspended Target Based on Single Camera

The low gravity test is an indispensable test for space manipulators before launching into the outer space. This paper aims to propose a position measurement method for the suspended target, which is a component of a low gravity test system. In order to obtain the position of lower hanging point in real time, an active target is designed with 7 infrared LEDs. A single camera is employed to acquire images of the feature points on the target. After that, the region growing, centroid extracting and locally searching methods are adopted to get the feature points location on the imagine plane. And then the perspective-n-point pose determination method (PnP) is used to calculate the relative position of the upper and lower hanging points. Experimental results show that this method can measure the horizontal position accurately and efficiently, and the maximum error is 0.128mm.

Robust Image Fingerprinting Based on Feature Point Relationship Mining

Local feature points have been widely employed in robust image fingerprinting. One of their intrinsic advantages is their invariance under geometric transforms. However, their robustness against certain attacks that modify the positions of points, such as additive noising and blurring, is limited. In addition, local-feature-point-based approaches ignore the distribution of the feature points. In this paper, we harness feature point relationships, including local structures and global relevance, to overcome these limitations. In the relationship mining strategy, Delaunay triangulation is first applied to the feature points to capture their geometric structures. Subsequently, local structures are represented by searching for an independent set in the mapping graph constructed via Delaunay triangulation, whereas the global relevance is represented by the Laplacian of the graph. Finally, the local structures and global relevance are used as input to the quantization process of the image fingerprinting system. In the process of quantization, we propose an unsupervised quantization strategy called between-cluster distance-based quantization to preserve the neighborhood structure between the binary fingerprint space and the original feature space. Experimental results show that the proposed method achieves effective performance under common modifications.

Robust feature point detectors for car make recognition

An Automatic Vehicle Make and Model Recognition (AVMMR) system can be a useful add-on tool to Automatic Number Plate Recognition (ANPR) to address potential car cloning, including intelligence collection by the police to outline past and recent car movement and travel patterns. Although several AVMMR systems have been proposed, the approaches perform sub-optimally under various environmental conditions, including occlusion and/or poor lighting distortions. This paper studies the effectiveness of deploying robust local feature points that can address these limitations. The proposed methods utilize a modification of two-dimensional feature points such as SIFT, SURF, etc. and their combinations. When SIFT gets combined with the multi-scale Harris/multi-scale Hessian methods, it could outperform existing approaches. Experimental evaluations using 4 different benchmark datasets are conducted to demonstrate the robustness of the proposed techniques and their abilities to detect and identify car makes and models under various environmental conditions. SIFT- DoG, SIFT- multiscale Hessian, and SIFT- multiscale Harris are shown to yield the best results for our datasets with higher recognition rates than those achieved with other existing methods in the literature. Therefore, it can then be concluded that the combination of certain covariant feature detectors and descriptors can outperform other methods.