Rotation invariant HOG for object localization in web images

The problem of detecting objects in images is solved by training the classifier by the descriptors constructed based on the local Radon transform of the gradient field of the image. The space of the Radon transform is considered as the Hough space accumulator in which projections are constructed. The set of local projections forms a descriptor of the region related to the object that is a generalization of the known histogram of oriented gradients (HOG) descriptor. The issues of the effect of the approximation of the Radon transform contribution function, the form of local normalization, and the number of directions in the projection histograms on the results of detecting pedestrians are investigated. The results produced by the proposed descriptor are compared with the results obtained using the HOG descriptor and convolutional neural networks (CNN) based on the ResNext and MobileNet architectures on the INRIA and CityScapes databases.

We present a Histogram of Oriented Gradient (HoG) based two-directional Dynamic Time Warping (DTW) matching method for handwritten word spotting. Firstly, we extract HoG descriptors from each cell in the normalized images. Then we connect the HoG descriptors in the same column and get a sequence of feature vectors. We do the same operation for the HoG descriptors in the same row. We then apply the two-directional DTW method to calculate the distance between the feature vectors sequences extracted from the query word and the candidate one. The experimental results show that the two-directional DTW is more robust to word deformation than the traditional DTW. And the local features such as HoG, LBP and SIFT combined with the two-directional DTW method outperform the method using the local feature descriptors directly. The HoG based two-directional DTW get the highest mean average precision on both the George Washington dataset and the CASIA-HWDB 2.1 dataset.

The Histogram of Oriented Gradients (HOG) is a popular feature descriptor used in computer vision and image processing. The technique counts occurrences of gradient orientation in localized portions of an image. The descriptor is sensible to the presence in images of noise, nonuniform illumination, and low contrast. In this work, we propose a robust HOG-based descriptor using the local energy model and phase congruency approach. Computer simulation results are presented for recognition of objects in images affected by additive noise, nonuniform illumination, and geometric distortions using the proposed and conventional HOG descriptors.

Maximum Curvature Method (MCM) is one of the promising methods for finger vein verification. MCM scans the curvature of the vein image profiles within a finger for feature extraction. However, the quality of the image can be poor due to variations in illumination and sensor conditions. Furthermore, traditional MCM matching of the vein pattern requires extensive processing time. To address these limitations, we propose an integrated Enhanced Maximum Curvature (EMC) method with Histogram of Oriented Gradient (HOG) descriptor for finger vein verification. Unlike MCM, EMC incorporates an enhancement mechanism to extract small vein delineation that is hardly visible in the extracted vein patterns. Next, HOG is applied instead of image binarization to convert a two-dimensional vein image into a one-dimensional feature vector for efficient matching. The HOG descriptor is able to characterize the local spatial representation of a finger vein by capturing the gradient information effectively. The proposed method is evaluated based on two datasets namely the PKU Finger Vein Database (V4) and SDUMLA-HMT finger vein database. Experiments show promising verification results with equal error rates as low as 0.33 % for DB1 and 0.14 % for DB2 respectively, when EMC+HOG+SVM is applied.

Face recognition is widely used in computer vision and in many other biometric applications where security is a major concern. The most common problem in recognizing a face arises due to pose variations, different illumination conditions and so on. The main focus of this paper is to recognize whether a given face input corresponds to a registered person in the database. Face recognition is done using Histogram of Oriented Gradients (HOG) technique in AT & T database with an inclusion of a real time subject to evaluate the performance of the algorithm. The feature vectors generated by HOG descriptor are used to train Support Vector Machines (SVM) and results are verified against a given test input. The proposed method checks whether a test image in different pose and lighting conditions is matched correctly with trained images of the facial database. The results of the proposed approach show minimal false positives and improved detection accuracy.

This paper presents a spatial layout recovery approach from single omnidirectional images. Vertical structures in the scene are extracted via classification from heterogeneous features computed at the superpixel level. Vertical surfaces are further classified according to their main orientation by fusing oriented line features, floor-wall boundary features and histogram of oriented gradients (HOG) with a Random Forest classifier. Oriented line features are used to build an orientation map that considers the main vanishing points. The floor-wall boundary feature attempts to reconstruct the scene shape as if it were observed from a bird's-eye view. Finally, the HOG descriptors are aggregated per superpixel and summarize the gradient distribution at homogeneous appearance regions. Compared to existing methods in the literature which rely only on corners or lines, our method gains statistical support from multiple cues aggregated per superpixel which provide more robustness against noise, occlusion, and clutter.

Face Recognition plays a very important role in numerous occasions based on visual security in recent days. The current methods of face recognition are to extract the different features of different faces to distinguish from others, so feature extraction has become a vital step in face recognition. For this reason, this paper presents a new fusion of local difference binary (LDB) and histogram of oriented gradients (HOG) for face recognition. We use LDB descriptor to extract the local pattern features of a face image. At the same time, the edge features of the original image are extracted by using HOG descriptor. The proposed new fusion of features improves the shortcomings of the low accuracy and avoids the problems that the dimension of the general fusion of features is too high. The experimental results on ORL and Yale face database verify the validity of the proposed fusion of features.

The aim of our work is to investigate the use of two powerful feature descriptor known as Zernike moments and histogram of oriented gradient (HOG) for facial images extracted from a video sequence uttered ten times by four speakers where MLP neural network classifier is used to identify the correct speaker. We've demonstrated good recognition rate obtained for both Zernike moments and HOG descriptors.

Semi-autonomous operation of intelligent vehicles may require that such platforms maintain a basic situational awareness with respect to people, other vehicles and their intent. These vehicles should be able to operate safely among people and other vehicles, and be able to perceive threats and respond accordingly. A key requirement is the ability to detect people and vehicles from a moving platform. We have developed one such algorithm using video cameras mounted on the vehicle. Our person detection algorithms model the shape and appearance of the person instead of modeling the background. This algorithm uses histogram of oriented gradients (HOG), which model shape and appearance using image edge histograms. These HOG descriptors are computed on an exhaustive set of image windows, which are then classified as person/non-person using a support vector machine classifier. The image windows are computed using camera calibration, which provides approximate size of people with respect to their location in the imagery. The algorithm is flexible and has been trained for different domains such as urban, rural and wooded scenes. We have designed a sensor platform that can be mounted on a moving vehicle to collect video data of pedestrians. Using manually annotated ground-truth data we have evaluated the person detection algorithm in terms of true positive and false positive rates. This paper provides a detailed overview of the algorithm, describes the experiments conducted and reports on algorithmic performance.

Of great importance are the tasks of recognizing objects of a given shape and color on complex background in conditions of interference with high accuracy. In this regard, the development of effective algorithms for selecting objects that reduce the influence of noise component and reduce computational costs of performing operations in which the result will be suitable for subsequent analysis within the framework of the tasks being solved is an urgent task. This paper presents a model of objects and a method based on radial symmetry transformation for the problems of recognizing round-shaped colored objects in color images. Examples of such objects include the pupils of the eyes, traffic lights, car lights and light markers. The aim of the work is to increase the accuracy of detecting round-shaped colored objects in the images under interference conditions using modified radial symmetry transformation. The paper describes a model and a method for detecting objects in images, presents the results of numerical experiments that allow us to estimate the limits of applicability of the proposed method.

Motivated by the recent success of supervised and weakly supervised common object discovery, in this paper, we move forward one step further to tackle common object discovery in a fully unsupervised way. Generally, object co-localization aims at simultaneously localizing objects of the same class across a group of images. Traditional object localization/detection usually trains specific object detectors which require bounding box annotations of object instances, or at least image-level labels to indicate the presence/absence of objects in an image. Given a collection of images without any annotations, our proposed fully unsupervised method is to simultaneously discover images that contain common objects and also localize common objects in corresponding images. Without requiring to know the total number of common objects, we formulate this unsupervised object discovery as a sub-graph mining problem from a weighted graph of object proposals, where nodes correspond to object proposals, and edges represent the similarities between neighbouring proposals. The positive images and common objects are jointly discovered by finding sub-graphs of strongly connected nodes, with each sub-graph capturing one object pattern. The optimization problem can be efficiently solved by our proposed maximal-flow-based algorithm. Instead of assuming that each image contains only one common object, our proposed solution can better address wild images where each image may contain multiple common objects or even no common object. Moreover, our proposed method can be easily tailored to the task of image retrieval in which the nodes correspond to the similarity between query and reference images. Extensive experiments on PASCAL VOC 2007 and Object Discovery data sets demonstrate that even without any supervision, our approach can discover/localize common objects of various classes in the presence of scale, view point, appearance variation, and partial occlusions. We also conduct broad experiments on image retrieval benchmarks, Holidays and Oxford5k data sets, to show that our proposed method, which considers both the similarity between query and reference images and also similarities among reference images, can help to improve the retrieval results significantly.

This paper presents a new algorithm to better classify objects in videos. In our case, the objects are cars, vans, and people on the roads. First, in order to extract the moving objects more precisely, we have proposed a method for foreground extraction based on the contour differences between the video frame and the background image. Second, after we got the integrated moving object, we have proposed a new algorithm to extract better features from the object. The new algorithm is based on two extended Histogram of Oriented Gradient (HOG) descriptor. We have improved HOG in two aspects: (a) selecting the gradient information from the moving objects and discarding the background gradient; (b) weighting every bin of gradient orientation histogram according to their significance within predefined area, in order to emphasize the important gradient information. We obtained Contour-Difference HOG (CD-HOG) from the first extension and Local-Main-Gradient-Orientation HOG (LMGO-HOG) from the second extended HOG. These extensions can cope with the cluttered background and make the features more distinguishable. Each of the extended HOG descriptors can produce a satisfying performance separately and an even better one if they are applied in cascade. From extensive evaluations, we showed the wonderful performance of our algorithm, and the accuracy rate of 94.04% can be achieved in some cases.

In this paper, we present a comparative evaluation of several appearance and shape descriptors in the context of 3D human pose estimation. Among the shape descriptors, we evaluate the Discrete Cosine Transform (DCT) and the Histogram of Shape Context (HoSC) descriptors. The five appearance descriptors that we evaluate are all variants of the Histogram of Oriented Gradients (HOG) descriptor. We evaluate these descriptors quantitatively using the HumanEva-I dataset. We report the performance of the descriptors using the Relevance Vector Machine (RVM) regression and K-nearest neighbor (KNN) regression methods. We found that the appearance descriptor computed at multiple spatial regions gave the best performance when RVM regression was used for pose estimation. The DCT descriptor performed the best when KNN regression was used for pose estimation.

The efficient separation of coal and gangue in the mining process is of great significance for improving coal mining efficiency and reducing environmental pollution. Automatic detection of coal and gangue is the key and foundation for the separation of coal and gangue. In this paper, we proposed a hierarchical framework for coal and gangue detection based on deep learning models. In this framework, the Gaussian pyramid principle is first used to construct multi-level training data, leading to the sets of coal and gangue image features with multiple scales. Then, the coal and gangue regional proposal networks (CG-RPN) are designed to determine the candidate regions of the target objects in the image. Next, convolution neural networks (CNNs) are constructed to recognize coal and gangue objects on the basis of extracted candidate regions. We performed our method on three different datasets. Experimental results showed that the proposed method improves the detection accuracy of coal and gangue objects by 0.8% compared with the previous methods, reaching up to 98.33%. In addition, our proposed method enables the detection of multiple coal and gangue objects in an individual image and solves the problem of queuing requirements in traditional methods.

This paper proposes a new detection and recognition method for moving objects that uses the temporal difference method (TDM) and the hidden Markov model (HMM). First, we apply the concept of entropy to convert the pixel value in the image domain into the amount of energy change in the entropy domain. Second, we use the temporal difference method to quickly detect the region of moving objects in complex images to address the variation in changing environments. Third, we use the discrete wavelet transformation technique to extract proper feature vectors from the detected mask image. Fourth, we use the hidden Markov model to accurately recognize moving objects. The results indicate that our proposed method can effectively and accurately detect and recognize moving objects in image sequences.