Invariant Shape Features Research Articles

Cascaded multi-3D-view fusion for 3D-oriented object detection

Currently, multi-view fusion methods fuse point- or proposal-level features from different views at the end stage of the backbone. This once-end-fusion method is not conducive to the timely adjustment of spatial misalignment for features from different views. Consequently, the discriminative depth and orientation details of the 3D oriented point cloud object may be filtered. To enhance the feature capture capability of the network, we introduce a cascaded multi-3D-view fusion method (CM3DV) to learn the implicit representation of object orientation. In particular, the proposed CM3DV method incorporates the cylindrical front view projection into a voxelised 3D bird’s-eye-view representation in a cascaded manner, and vice versa. Through the learning of 3D-regulated instance representation, this bi-directional mutual fusion module, called cascaded multi-view feature fusion module, alleviates the spatial misalignment of the two views. Furthermore, to learn the rotation- and shape-invariant features of objects, modulated rotation head (MRH) develops a direction-guided adjustment instead of an axis-aligned structure to extract instance-consistent features. By excluding the irrelevant content using MRH, this instance-consistent feature will benefit the object classification and orientation regression. Extensive experiments on the KITTI dataset show that the proposed method achieves a significant improvement over existing advanced methods, especially for orientation estimation.

Global and local shape features of the hippocampus based on Laplace-Beltrami eigenvalues and eigenfunctions: a potential application in the lateralization of temporal lobe epilepsy.

Using magnetic resonance (MR) images to evaluate changes in the shape of the hippocampus has been an active research topic. This paper presents a new shape analysis approach to quantify and visualize deformations of the hippocampus in epilepsy. The proposed method is based on Laplace-Beltrami (LB) eigenvalues and eigenfunctions as isometric invariant shape features, and thus, the procedure does not require any image registration. In addition to the LB-based shape features, total hippocampal volume and surface area are calculated using manually segmented images. Theses shape and volumetric descriptors are used to distinguish the patients with temporal lobe epilepsy (TLE) (N = 55) from healthy control subjects (N = 12, age = 32.2 ± 9.1, sex (M/F) = 6/6) and patients with right TLE (N = 26, age = 45.1 ± 11.0, sex (M/F) = 9/17) from left TLE (N = 29, age = 45.4 ± 11.9, sex (M/F) = 10/19). Experimental results illustrate the usefulness of the proposed approach for the diagnosis and lateralization of TLE with 93.0% and 86.4% of the cases, respectively. Moreover, the proposed method outperforms the volumetric analysis in terms of both sensitivity (94.9% vs. 88.1%) and specificity (83.3% vs. 50.0%) of the lateralization. The analysis of local hippocampal thickness variations suggests significant deformation in both ipsilateral and contralateral hippocampi of epileptic patients, while there were no differences between right and left hippocampi in controls. It is anticipated that the proposed method could be advantageous in the presurgical evaluation of patients with drug-resistant epilepsy; however, further validation of the method using a larger dataset is required.

Content-Based Image Retrieval Using Colour, Gray, Advanced Texture, Shape Features, and Random Forest Classifier with Optimized Particle Swarm Optimization.

In this paper, a new approach for Content-Based Image Retrieval (CBIR) has been addressed by extracting colour, gray, advanced texture, and shape features for input query images. Contour-based shape feature extraction methods and image moment extraction techniques are used to extract the shape features and shape invariant features. The informative features are selected from extracted features and combined colour, gray, texture, and shape features by using PSO. The target image has been retrieved for the given query image by training the random forest classifier. The proposed colour, gray, advanced texture, shape feature, and random forest classifier with optimized PSO (CGATSFRFOPSO) provide efficient retrieval of images in a large-scale database. The main objective of this research work is to improve the efficiency and effectiveness of the CBIR system by extracting the features like colour, gray, texture, and shape from database images and query images. These extracted features are processed in various levels like removing redundancy by optimal feature selection and fusion by optimal weighted linear combination. The Particle Swarm Optimization algorithm is used for selecting the informative features from gray and colour and texture features. The matching accuracy and the speed of image retrieval are improved by an ensemble of machine learning algorithms for the similarity search.

Efficient retrieval of trademark images from large databases

Trademarks law is considered as the most pervasive amongst all the intellectual property laws because all the judgments require contemplating the imaginations of the consumers. Once the trademark has been submitted for registration, the examiners at the trademark office make sure that it is not similar to any of the previous registered trademarks. This motivated the need of an automated trademark retrieval system. This paper makes a contribution in the field of trademark image retrieval by proposing a retrieval technique that allows a flexible combination of colour, texture and shape features. Moreover the proposed technique utilises HSV colour histogram, for colour, multi resolution Gabor wavelet for texture and an integration of Zernike moments for global shape and scale invariant feature transform (SIFT) for local shape feature extraction. The results have been tested on MPEG7, MPEG trademark, WANG and self-compiled datasets. The improvement achieved in precision is 14% on MPEG7, 30% on the WANG and 38% on self-compiled dataset. Similarly, 26% improvement in average recall is achieved and around 16% when our proposed shape feature is compared with other state of the art techniques like Fourier descriptors, Hu moments, wavelet descriptors, Zernike moments and edge gradient co-occurrence matrix.

RBC Classification in Blood Smear Image using Neural Network

Biomedical image processing becomes an emerging field due to automation in the field of medical science with the help of image processing techniques. In medical science it is very much essential to diagnosis a disease accurately and efficiently. Most of the disease which deals with the blood test report for diagnosis of the disease. This paper proposed a computer vision based method which extract the Red Blood Cells (RBC) from a blood smear image and classify it whether normal or abnormal. Then it will count the normal RBC as well as abnormal RBC. This method works in two parts, one is segmentation of blood cell and other is classification and counting of segmented blood cells using neural network. The Neural network trained and classified using shape and moment invariant features because this features are invariant to translation, scaling and rotation. The proposed method performs well and gives about 90 percent of correct result.

Ship Detection in Panchromatic Optical Remote Sensing Images Based on Visual Saliency and Multi-Dimensional Feature Description

Ship detection in panchromatic optical remote sensing images is faced with two major challenges, locating candidate regions from complex backgrounds quickly and describing ships effectively to reduce false alarms. Here, a practical method was proposed to solve these issues. Firstly, we constructed a novel visual saliency detection method based on a hyper-complex Fourier transform of a quaternion to locate regions of interest (ROIs), which can improve the accuracy of the subsequent discrimination process for panchromatic images, compared with the phase spectrum quaternary Fourier transform (PQFT) method. In addition, the Gaussian filtering of different scales was performed on the transformed result to synthesize the best saliency map. An adaptive method based on GrabCut was then used for binary segmentation to extract candidate positions. With respect to the discrimination stage, a rotation-invariant modified local binary pattern (LBP) description was achieved by combining shape, texture, and moment invariant features to describe the ship targets more powerfully. Finally, the false alarms were eliminated through SVM training. The experimental results on panchromatic optical remote sensing images demonstrated that the presented saliency model under various indicators is superior, and the proposed ship detection method is accurate and fast with high robustness, based on detailed comparisons to existing efforts.

ARMBIS: accurate and robust matching of brain image sequences from multiple modal imaging techniques.

Study of brain images of rodent animals is the most straightforward way to understand brain functions and neural basis of physiological functions. An important step in brain image analysis is to precisely assign signal labels to specified brain regions through matching brain images to standardized brain reference atlases. However, no significant effort has been made to match different types of brain images to atlas images due to influence of artifact operation during slice preparation, relatively low resolution of images and large structural variations in individual brains. In this study, we develop a novel image sequence matching procedure, termed accurate and robust matching brain image sequences (ARMBIS), to match brain image sequences to established atlas image sequences. First, for a given query image sequence a scaling factor is estimated to match a reference image sequence by a curve fitting algorithm based on geometric features. Then, the texture features as well as the scale and rotation invariant shape features are extracted, and a dynamic programming-based procedure is designed to select optimal image subsequences. Finally, a hierarchical decision approach is employed to find the best matched subsequence using regional textures. Our simulation studies show that ARMBIS is effective and robust to image deformations such as linear or non-linear scaling, 2D or 3D rotations, tissue tear and tissue loss. We demonstrate the superior performance of ARMBIS on three types of brain images including magnetic resonance imaging, mCherry with 4',6-diamidino-2-phenylindole (DAPI) staining and green fluorescent protein without DAPI staining images. The R software package is freely available at https://www.synapse.org/#!Synapse:syn18638510/wiki/591054 for Not-For-Profit Institutions. If you are a For-Profit Institution, please contact the corresponding author. Supplementary data are available at Bioinformatics online.

Geometrically invariant color, shape and texture features for object recognition using multiple kernel learning classification approach

The geometrically invariant image descriptors are very important for recognizing objects of arbitrary shapes and orientations. We propose a framework for the fusion of the geometrically invariant descriptors representing color, shape, and texture for the recognition of color objects using multiple kernel learning (MKL) approach. To describe texture of color images, we propose an effective rotation invariant texture descriptor which is based on the Zernike moments (ZMs) of the gradient of the color images, referred to as the GZMs. For the shape features, we use the ZMs of the intensity component of a color image and also use multi-channel ZMs (MZMs) which have proven to be superior in performance than the quaternion ZMs (QZMs). For comparative performance analysis, rotation invariants of the QZMs (RQZMs) are also considered. Since the color histograms (CH) are known to be very effective color descriptors, we consider them for representing color. The five sets of features – CH, ZMs, GZMs, MZMs, and RQZMs are invariant to translation, rotation, and scale. The fusion of color, shape and texture features in different combinations using the MKL approach is shown to provide very high recognition rates on PASCAL VOC 2005, Soccer, SIMPLIcity, Flower, and Caltech-101 datasets.

RGBD Human Action Recognition using Multi-Features Combination and K-Nearest Neighbors Classification

In this paper, we present a novel system to analyze human body motions for action recognition task from two sets of features using RGBD videos. The Bag-of-Features approach is used for recognizing human action by extracting local spatialtemporal features and shape invariant features from all video frames. These feature vectors are computed in four steps: Firstly, detecting all interest keypoints from RGB video frames using Speed-Up Robust Features and filters motion points using Motion History Image and Optical Flow, then aligned these motion points to the depth frame sequences. Secondly, using a Histogram of orientation gradient descriptor for computing the features vector around these points from both RGB and depth channels, then combined these feature values in one RGBD feature vector. Thirdly, computing Hu-Moment shape features from RGBD frames; fourthly, combining the HOG features with Hu-moments features in one feature vector for each video action. Finally, the k-means clustering and the multi-class K-Nearest Neighbor is used for the classification task. This system is invariant to scale, rotation, translation, and illumination. All tested, are utilized on a dataset that is available to the public and used often in the community. By using this new feature combination method improves performance on actions with low movement and reach recognition rates superior to other publications of the dataset.

Joint feature extraction technique for text detection from natural scene image

Detection text detection and extraction from natural scenes (i.e. video or images) can deliver significant information for various applications. To address the issue of text detection, a novel approach for text detection from natural scene image is introduced by developing a joint feature extraction method by considering shape and scale invariant feature transform (SIFT) feature analysis techniques. Shape extraction is improved by applying curvature-based shape analysis model. To construct the feature descriptor, input image is passed through canny edge detection process in which gradients are computed of each image. Later, we perform SIFT analysis and SIFT-based feature matching to formulate the SIFT feature descriptor. Finally, these two descriptors are merged together, and a combined descriptor is presented for text detection. Experimental study is carried out by considering benchmark ICDAR 2003, 2013 and 2015 data sets. Experimental study shows that proposed approach outperforms when compared with state-of-art text detection model.

A Lesion-Based Response Prediction Model Using Pretherapy PET/CT Image Features for Y90 Radioembolization to Hepatic Malignancies

We present a probabilistic approach to identify patients with primary and secondary hepatic malignancies as responders or nonresponders to yttrium-90 radioembolization therapy. Recent advances in computer-aided detection have decreased false-negative and false-positive rates of perceived abnormalities; however, there is limited research in using similar concepts to predict treatment response. Our approach is driven by the goal of precision medicine to determine pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography imaging parameters to facilitate the identification of patients who would benefit most from yttrium-90 radioembolization therapy, while avoiding complex and costly procedures for those who would not. Our algorithm seeks to predict a patient's response by discovering common co-occurring image patterns in the lesions of baseline fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography scans by extracting invariant shape and texture features. The extracted imaging features were represented as a distribution of each subject based on the bag-of-feature paradigm. The distribution was applied in a multinomial naive Bayes classifier to predict whether a patient would be a responder or nonresponder to yttrium-90 radioembolization therapy based on the imaging features of a pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography scan. Comprehensive published criteria were used to determine lesion-based clinical treatment response based on fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography imaging findings. Our results show that the model is able to predict a patient with liver cancer as a responder or nonresponder to yttrium-90 radioembolization therapy with a sensitivity of 0.791 using extracted invariant imaging features from the pretherapy fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography test. The sensitivity increased to 0.821 when combining extracted invariant image features with variable features of tumor volume.

A Vocabulary Approach to Partial Streamline Matching and Exploratory Flow Visualization.

Measuring the similarity of integral curves is fundamental to many important flow data analysis and visualization tasks such as feature detection, pattern querying, streamline clustering, and hierarchical exploration. In this paper, we introduce FlowString, a novel vocabulary approach that extracts shape invariant features from streamlines and utilizes a string-based method for exploratory streamline analysis and visualization. Our solution first resamples streamlines by considering their local feature scales. We then classify resampled points along streamlines based on the shape similarity around their local neighborhoods. We encode each streamline into a string of well-selected shape characters, from which we construct meaningful words for querying and retrieval. A unique feature of our approach is that it captures intrinsic streamline similarity that is invariant under translation, rotation and scaling. We design an intuitive interface and user interactions to support flexible querying, allowing exact and approximate searches for partial streamline matching. Users can perform queries at either the character level or the word level, and define their own characters or words conveniently for customized search. We demonstrate the effectiveness of FlowString with several flow field data sets of different sizes and characteristics. We also extend FlowString to handle multiple data sets and perform an empirical expert evaluation to confirm the usefulness of this approach.

Hand Gesture Recognition and Its Application in Robot Control

In view of the problem that the accuracy and robustness of hand gesture recognition technology based on vision in the process of using gestures to interact with robots were unstable due to its background, illumination and other factors, this paper presented a gesture segmentation and recognition method Combining depth information and color images. First, it used Kinect sensor to obtain depth information and color images, then used depth information to pick the hand part from color images, and then got gesture images through color segmentation method. Second, it calculated HU invariant moments and shape features of the gesture images as feature information. Finally, it used the feature information to train the support vector machine, then implemented hand gesture recognition for static hand gestures. Experimental results show that the method has strong robustness to the influence of background interference, illumination variation, translation, rotation and zoom, and can be applied to control intelligent robot.

Shaping diffraction-free Lommel beams with digital binary amplitude masks.

Here, we report on experimental observations of various Lommel modes that possess distinct diffraction-free behaviors. The binary amplitude masks are designed to accurately encode the complex field information with the superpixel method. Then, the generation of the desired beams is demonstrated with these binary patterns projected onto the digital micromirrors device (DMD). Remarkably, we find that the field distribution and orbital angular momentum can be continuously engineered by tuning the beam parameters. Furthermore, the shape-invariant feature of such beams is verified by their far-field ring-like structures. Our observations are in good accordance with the theoretical predictions, and our methods may find potential applications in optical guiding and imaging. Moreover, apart from the DMD, the binary amplitude hologram can also be presented with well-fabricated elements, and thus, our method will also enable new applications for surface plasmon polaritons as well as electron beams.

Multi-view object instance recognition in an industrial context

SUMMARYWe present a fast object recognition system coding shape by viewpoint invariant geometric relations and appearance information. In our advanced industrial work-cell, the system can observe the work space of the robot by three pairs of Kinect and stereo cameras allowing for reliable and complete object information. From these sensors, we derive global viewpoint invariant shape features and robust color features making use of color normalization techniques.We show that in such a set-up, our system can achieve high performance already with a very low number of training samples, which is crucial for user acceptance and that the use of multiple views is crucial for performance. This indicates that our approach can be used in controlled but realistic industrial contexts that require—besides high reliability—fast processing and an intuitive and easy use at the end-user side.

평균 형상 모델과 SIFT 특징을 이용한 TRUS 영상의 전립선 분할

전립선암은 남자에게 가장 흔히 나타나는 암 중의 하나이며, 많은 나라에서 죽음에 이르게 하는 큰 요인이 되고 있다. 전립선암을 진단하고 치료하는 과정에서 비용이 싼 TRUS 영상이 사용된다. 그러나 전립선 경계의 정확한 구분이 요구되지만 어려운 문제이다. 그 이유는 경계가 불명확하고, 반점들이 많으며, 그레이 레벨의 범위가 작기 때문이다. 본 연구에서는 전립선의 평균 형상 모델과 불변의 특징을 이용하여 TRUS 영상에서 자동으로 전립선 분할하는 방법을 제안한다. 이 방법은 4 단계로 구성된다. 먼저, 에지 분포를 이용하여 프로브와 두개의 직선을 찾아낸다. 다음으로, 평균 형상 모델의 중앙에 위치한 3개의 전립선 패치를 획득한다. 이 패치는 전립선과 비전립선의 특징을 비교하기 위해 사용된다. 다음으로, 세 개의 패치와 각 블록들이 얼마나 대표 블록과 유사한지를 비교한다. 마지막으로, 앞 단계의 경계와 첫 단계에서 얻은 개략적 경계가 최종 분할에 사용된다. 이 방법의 유효성을 검증하기 위하여 실험을 하였으며, 인간 전문가에 의해 얻어진 경계와 비교하여 7.78% 미만의 차이로 경계를 얻을 수 있었다. Prostate cancer is one of the most frequent cancers in men and is a major cause of mortality in the most of countries. In many diagnostic and treatment procedures for prostate disease, transrectal ultrasound(TRUS) images are being used because the cost is low. But, accurate detection of prostate boundaries is a challenging and difficult task due to weak prostate boundaries, speckle noises and the short range of gray levels. This paper proposes a method for automatic prostate segmentation in TRUS images using its average shape model and invariant features. This approach consists of 4 steps. First, it detects the probe position and the two straight lines connected to the probe using edge distribution. Next, it acquires 3 prostate patches which are in the middle of average model. The patches will be used to compare the features of prostate and nonprostate. Next, it compares and classifies which blocks are similar to 3 representative patches. Last, the boundaries from prior classification and the rough boundaries from first step are used to determine the segmentation. A number of experiments are conducted to validate this method and results showed that this new approach extracted the prostate boundary with less than 7.78% relative to boundary provided manually by experts.

A simple computational model for image retrieval with weighted multifeatures based on orthogonal polynomials and genetic algorithm

This paper proposes a simple and new image retrieval method with weighted multifeature set based on multiresolution enhanced orthogonal polynomials model and genetic algorithm. In the proposed method, initially the orthogonal polynomials model coefficients are computed and reordered into multiresolution subband like structure. Then the statistical, directional, perceptual and invariant texture, shape and color features are directly extracted from the subband coefficients. The extracted texture, shape and color features are integrated into linear multifeature set and the significance of each feature in the multifeature set is determined by assigning appropriate weight. This paper also proposes a method to compute the optimized weight for each feature in the integrated linear multifeature multi feature set using genetic algorithm. Then the obtained optimized weight is multiplied with the corresponding features in the multifeature set and the weighted Manhattan distance metric is used for retrieving similar images. The efficiency of the proposed method is experimented on the standard subset of Corel and Caltech database images. The performance of the proposed method is compared with other existing retrieval methods such as Haar wavelet and Contourlet Transform based retrieval schemes. The proposed method yields high average recall and precision of 92.6% and 71% for Corel database and 90.5% and 72.3% of Caltech database images when compared with other existing methods.

Hierarchical human activity recognition system based on R-transform and nonlinear kernel discriminant features

A framework for a video based hierarchical human activity recognition (HAR) system is presented based on efficient feature extraction and dimension reduction techniques R-transform and kernel discriminant analysis (KDA). The hierarchical HAR system is proposed to group similar activities and further improve the recognition rate. A first level system uses R-transform to extract symmetric, scale and translation invariant shape features from the silhouette sequences and KDA is applied on the R-transformed features to increase discrimination among different classes of activities based on their nonlinear representations from different view angles. A second level system is applied selectively to the recognised activities from the first level system to increase further discrimination for the activities with high similarity in postures. The system is validated with a recognition rate of 97.3% for the KTH dataset and 99.1% for the Weizmann dataset. The improved recognition rate for the hierarchical HAR system compared to state of the art on the KTH and Weizmann datasets demonstrates the effectiveness of the proposed system.

A double mapping framework for extraction of shape-invariant features based on multi-scale partitions with AdaBoost for video smoke detection

Traditional methods for video smoke detection can easily achieve very low training errors but their generalization performances are not good due to arbitrary shapes of smoke, intra-class variations, occlusions and clutters. To overcome these problems, a double mapping framework is proposed to extract partition based features with AdaBoost. The first mapping is from an original image to block features. A feature vector is presented by concatenating histograms of edge orientation, edge magnitude and Local Binary Pattern (LBP) bit, and densities of edge magnitude, LBP bit, color intensity and saturation. Each component of the feature vector produces a feature image. To obtain shape-invariant features, a detection window is partitioned into a set of small blocks called a partition, and many multi-scale partitions are generated by changing block sizes and partition schemes. The sum of each feature image within each block of each partition is computed to generate block features. The second mapping is from the block features to statistical features. The statistical features of the block features, such as, mean, variance, skewness, kurtosis and Hu moments, are computed on all partitions to form a feature pool. AdaBoost is used to select discriminative shape-invariant features from the feature pool. Experiments show that the proposed method has better generalization performance and less insensitivity to geometry transform than traditional methods.

Object Recognition Technique based on Level Set Method and Neural Network

The Object recognition is the task of finding and labeling parts of a two-dimensional (2D) image of a scene that correspond to objects in the scene. In this paper, we have proposed an efficient approach using level set method for extracting object shape contour and convex hull as a shape invariant features to the Feed forward Neural Network classifier for object recognition. We extracted the shape contour by level set method. Then, we have obtained invariant shape feature, convex hull of the objects. This convex hull set serves as a pattern for the Neural Network. Initially Feed forward neural network trained on the odd data set and tested on even data set. Our approach is evaluated on the Amsterdam Library of Object Images collection, a large collection of object images containing 1000 objects recorded under various imaging circumstances. The experimental results clearly demonstrate that our approach significantly outperforms. The proposed method is shown to be effective under a wide variety of imaging conditions.