Framework For Object Recognition Research Articles

Saliency Driven Object recognition in egocentric videos with deep CNN: toward application in assistance to Neuroprostheses

The problem of object recognition in natural scenes has been recently successfully addressed with Deep Convolutional Neuronal Networks giving a significant break-through in recognition scores. The computational efficiency of Deep CNNs as a function of their depth, allows for their use in real-time applications. One of the key issues here is to reduce the number of windows selected from images to be submitted to a Deep CNN. This is usually solved by preliminary segmentation and selection of specific windows, having outstanding “objectiveness” or other value of indicators of possible location of objects. In this paper we propose a Deep CNN approach and the general framework for recognition of objects in a real-time scenario and in an egocentric perspective. Here the window of interest is built on the basis of visual attention map computed over gaze fixations measured by a glass-worn eye-tracker. The application of this set-up is an interactive user-friendly environment for upper-limb amputees. Vision has to help the subject to control his worn neuro-prosthesis in case of a small amount of remaining muscles when the EMG control becomes inefficient. The recognition results on a specifically recorded corpus of 151 videos with simple geometrical objects show the mean Average Precision (mAP) of 64,6% and the computational time at the generalization lower than a time of a visual fixation on the object of interest.

A Computational Visual Neuroscience Model for Object Recognition

In this study with the inspirations from both neuroscience and computer science, a combinatorial framework for object recognition was proposed having benefited from the advantages of both biologically-inspired HMAX_S architecture model for feature extraction and Extreme Learning Machine (ELM) as a classifier. HMAX model is a feed-forward hierarchical structure resembling the ventral pathway in the visual cortex of the brain and ELM is a powerful neural network, which randomly chooses hidden nodes and specifies analytically the single-hidden layer. ELM theories conjecture that this randomness may be true for biological learning in animal brains. It should be noted that the principle reason of using ELM is mainly as a result of its biological structure in order to imitate the biological object recognition system of mammalians and partly for its incredible speed which drastically lessens the runtime. Classification results are reported in Caltech101 dataset, at the focal point with its combinatorial framework serving considerable improvements over latest studies in both classification rate (96.39%) and the low runtime (0.417s).

Efficient graph-based search for object detection

In this paper, we focus on the object localization problem in images given a single hand-drawn example as the object model. We propose a novel framework for shape-based object detection and recognition, which we formulate as a graph-based search problem. In our method, we first propose five preprocessing procedures to reduce the irrelevant edge fragments in cluttered real images that often occur in edge detection. Then we build a graph to represent the edge images. Therefore, our goal is to find the group of adjacent nodes in the graph that match well to the model contours. Finally, we present a new evaluation method to verify the candidate hypotheses. We did experiments on the ETHZ shape classes dataset and the INRIA horses dataset. Experimental results demonstrate that the proposed method achieves not only accurate object detection but also precise contour localization in cluttered real images. Comparisons with other recent template-based matching methods further demonstrate the effectiveness and efficiency of the proposed method.

Extraction of sparse features of color images in recognizing objects

In this paper, we propose a new object recognition framework that combines Gabor energy filters, a visual cortex model in which single units alternate with complex units, and color information. Each color image is first converted to the CIELAB color space. Rather using Gabor filters in the first layer of the cortex model, to each color component, a set of Gabor energy filters is applied. Thereafter, the superposition responses of the Gabor energy filter outputs over the color components are normalized by divisive normalization. In the fourth layer, sparse features are calculated using a localized pooling method that allows retention of some geometric information from the prototype patches’ positions. Finally, a set of sparse features are exploited by a linear SVM classifier for object recognition and classification. In the learning stage, a set of prototypes is selected randomly over spatial position, spatial size, and several scales simultaneously, and is extracted by the local maximum over scales and orientations, ignoring weaker training scales and orientations. The results of experiments performed on several datasets show that the use of color information in our framework improves object recognition significantly.

A Unified Factors Analysis Framework for Discriminative Feature Extraction and Object Recognition

Various methods for feature extraction and dimensionality reduction have been proposed in recent decades, including supervised and unsupervised methods and linear and nonlinear methods. Despite the different motivations of these methods, we present in this paper a general formulation known as factor analysis to unify them within a common framework. During factor analysis, an object can be seen as being comprised of content and style factors, and the objective of feature extraction and dimensionality reduction is to obtain the content factor without style factor. There are two vital steps in factor analysis framework; one is the design of factor separating objective function, including the design of partition and weight matrix, and the other is the design of space mapping function. In this paper, classical Linear Discriminant Analysis (LDA) and Locality Preserving Projection (LPP) algorithms are improved based on factor analysis framework, and LDA based on factor analysis (FA-LDA) and LPP based on factor analysis (FA-LPP) are proposed. Experimental results show the superiority of our proposed approach in classification performance compared to classical LDA and LPP algorithms.

A Feature Learning and Object Recognition Framework for Underwater Fish Images.

Live fish recognition is one of the most crucial elements of fisheries survey applications where the vast amount of data is rapidly acquired. Different from general scenarios, challenges to underwater image recognition are posted by poor image quality, uncontrolled objects and environment, and difficulty in acquiring representative samples. In addition, most existing feature extraction techniques are hindered from automation due to involving human supervision. Toward this end, we propose an underwater fish recognition framework that consists of a fully unsupervised feature learning technique and an error-resilient classifier. Object parts are initialized based on saliency and relaxation labeling to match object parts correctly. A non-rigid part model is then learned based on fitness, separation, and discrimination criteria. For the classifier, an unsupervised clustering approach generates a binary class hierarchy, where each node is a classifier. To exploit information from ambiguous images, the notion of partial classification is introduced to assign coarse labels by optimizing the benefit of indecision made by the classifier. Experiments show that the proposed framework achieves high accuracy on both public and self-collected underwater fish images with high uncertainty and class imbalance.

Robust object recognition via weakly supervised metric and template learning

In this paper, we present a new framework for object recognition via weakly supervised metric and template learning, wherein the optimal metric and templates are jointly learned. Its advantages include high computational speed, and robustness against image noise and unbalanced training data. Specifically, considering the noise in the training data, our framework is formulated as a weakly supervised learning model in which images with higher reliability will contribute more to the training result. A latent structural SVM based Weakly Supervised Metric and Template Learning (WSMTL) method is designed to jointly learn the metric, the templates, and a weight vector. The weight vector is used to represent each image׳s reliability. With the learned metric and object templates, each testing sample is recognized via 1-NN searching within templates. Owing to the 1-NN searching scheme in the recognition phase, WSMTL is of great computational efficiency. We used CMU PIE database with synthesized noise to evaluate the robustness of WSMTL. Experimental results show that WSMTL is robust against noise and unbalanced training data. Moreover, we compared it with some state-of-the-art recognition methods on the public traffic sign dataset BTSC and human face database, i.e., Extended Yale-B. The comparison results demonstrate that our method outperforms the others in object recognition tasks.

Large-Margin Multi-Modal Deep Learning for RGB-D Object Recognition

Most existing feature learning-based methods for RGB-D object recognition either combine RGB and depth data in an undifferentiated manner from the outset, or learn features from color and depth separately, which do not adequately exploit different characteristics of the two modalities or utilize the shared relationship between the modalities. In this paper, we propose a general CNN-based multi-modal learning framework for RGB-D object recognition. We first construct deep CNN layers for color and depth separately, which are then connected with a carefully designed multi-modal layer. This layer is designed to not only discover the most discriminative features for each modality, but is also able to harness the complementary relationship between the two modalities. The results of the multi-modal layer are back-propagated to update parameters of the CNN layers, and the multi-modal feature learning and the back-propagation are iteratively performed until convergence. Experimental results on two widely used RGB-D object datasets show that our method for general multi-modal learning achieves comparable performance to state-of-the-art methods specifically designed for RGB-D data.

A Global Hypothesis Verification Framework for 3D Object Recognition in Clutter.

Pipelines to recognize 3D objects despite clutter and occlusions usually end up with a final verification stage whereby recognition hypotheses are validated or dismissed based on how well they explain sensor measurements. Unlike previous work, we propose a Global Hypothesis Verification (GHV) approach which regards all hypotheses jointly so as to account for mutual interactions. GHV provides a principled framework to tackle the complexity of our visual world by leveraging on a plurality of recognition paradigms and cues. Accordingly, we present a 3D object recognition pipeline deploying both global and local 3D features as well as shape and color. Thereby, and facilitated by the robustness of the verification process, diverse object hypotheses can be gathered and weak hypotheses need not be suppressed too early to trade sensitivity for specificity. Experiments demonstrate the effectiveness of our proposal, which significantly improves over the state-of-art and attains ideal performance (no false negatives, no false positives) on three out of the six most relevant and challenging benchmark datasets.

What, Where and How? Introducing pose manifolds for industrial object manipulation

In this paper we propose a novel method for object grasping that aims to unify robot vision techniques for efficiently accomplishing the demanding task of autonomous object manipulation. Through ontological concepts, we establish three mutually complementary processes that lead to an integrated grasping system able to answer conjunctive queries such as “What”, “Where” and “How”? For each query, the appropriate module provides the necessary output based on ontological formalities. The “What” is handled by a state of the art object recognition framework. A novel 6 DoF object pose estimation technique, which entails a bunch-based architecture and a manifold modeling method, answers the “Where”. Last, “How” is addressed by an ontology-based semantic categorization enabling the sufficient mapping between visual stimuli and motor commands.

Design and Realization of Pre-Amplifier and Filters for on-Board Radar System

Today's warfare situation incorporates a considerable measure of innovative weaponry and every nation is attempting to outmaneuver the other utilizing high engineering, henceforth it is of most extreme criticalness to be at the front line of the bleeding edge innovation and having the best weapons conceivable to secure the country. One of the innovations which give the one having it the high ground is Radio Detection and Ranging (RADAR) engineering. Radar is an object recognition framework that uses electromagnetic waves to recognize the range, vehicles, climate developments and landscape. A radar framework incorporates a transmitter which radiates an arrangement of EM pulses. These pulses are transmitted in a directional pattern using a suitable high-gain antenna. The radar then uses a collector, normally joined with the same radio wire, to catch any returned pulses being reflected from items showing up in the reception apparatus' field of perspective. The transmitter, antenna and recipient are connected utilizing a circulator. This guarantees that all the transmitter power is sent to the radio wire whilst all the reflected power came back to the reception apparatus. The echo signal coming back to the receiver is a high recurrence signal in Ghz extent having low adequacy. The processing of this sort of signal is extremely troublesome and henceforth it needs to be down changed over into an Intermediate Frequency (IF) signal whose reach is in Mhz. This down transformation is carried out utilizing a super heterodyne receiver. Most radar receivers use megahertz Intermediate Frequency (IF) with a value around 30 and 75 megahertz. Ground based radar is utilized for missile tracking framework to get its range and angular position to block and destroy. Precision of these radars is not sufficient to hit and crush, so the rocket itself is provided with onboard radar to track the target, block it buzzing around and pulverize. Radar signal handling of onboard radar is carried out in both simple and advanced modes, The latter mode of signal preparing is much simple to actualize changes and up gradation. But the IF signal before digitization needs to be amplified to get enough signal strength and filtered to select the desired. This paper exhibits the design and realization of an amplifier and filters to process the received signal at IF level of a locally available radar framework. MULTISIM is utilized to confirm the channel qualities and intensification variable.

A robust SVM classification framework using PSM for multi-class recognition

Our research focuses on the question of classifiers that are capable of processing images rapidly and accurately without having to rely on a large-scale dataset, thus presenting a robust classification framework for both facial expression recognition (FER) and object recognition. The framework is based on support vector machines (SVMs) and employs three key approaches to enhance its robustness. First, it uses the perturbed subspace method (PSM) to extend the range of sample space for task sample training, which is an effective way to improve the robustness of a training system. Second, the framework adopts Speeded Up Robust Features (SURF) as features, which is more suitable for dealing with real-time situations. Third, it introduces region attributes to evaluate and revise the classification results based on SVMs. In this way, the classifying ability of SVMs can be improved.Combining these approaches, the proposed method has the following beneficial contributions. First, the efficiency of SVMs can be improved. Experiments show that the proposed approach is capable of reducing the number of samples effectively, resulting in an obvious reduction in training time. Second, the recognition accuracy is comparable to that of state-of-the-art algorithms. Third, its versatility is excellent, allowing it to be applied not only to object recognition but also FER.

VFM: Visual Feedback Model for Robust Object Recognition

Object recognition, which consists of classification and detection, has two important attributes for robustness: 1) closeness: detection windows should be as close to object locations as possible, and 2) adaptiveness: object matching should be adaptive to object variations within an object class. It is difficult to satisfy both attributes using traditional methods which consider classification and detection separately; thus recent studies propose to combine them based on confidence contextualization and foreground modeling. However, these combinations neglect feature saliency and object structure, and biological evidence suggests that the feature saliency and object structure can be important in guiding the recognition from low level to high level. In fact, object recognition originates in the mechanism of "what" and "where" pathways in human visual systems. More importantly, these pathways have feedback to each other and exchange useful information, which may improve closeness and adaptiveness. Inspired by the visual feedback, we propose a robust object recognition framework by designing a computational visual feedback model (VFM) between classification and detection. In the "what" feedback, the feature saliency from classification is exploited to rectify detection windows for better closeness; while in the "where" feedback, object parts from detection are used to match object structure for better adaptiveness. Experimental results show that the "what" and "where" feedback is effective to improve closeness and adaptiveness for object recognition, and encouraging improvements are obtained on the challenging PASCAL VOC 2007 dataset.

Generic Object Recognition with Local Receptive Fields Based Extreme Learning Machine

Abstract Generic object recognition is to classify the object to a generic category. Intra-class variabilities cause big troubles for this task. Traditional methods involve plenty of pre-processing steps, like model construction, feature extraction, etc. Moreover, these methods are only effective for some specific dataset. In this paper, we propose to use local receptive fields based extreme learning machine (ELM-LRF) as a general framework for object recognition. It is operated directly on the raw images and thus suitable for all different datasets. Additionally, the architecture is simple and only requires few computations, as most connection weights are randomly generated. Comparing to state-of-the-art results on NORB, ETH-80 and COIL datasets, it is on par with the best one on ETH-80 and sets the new records for NORB and COIL.

Using Scale Coordination and Semantic Information for Robust 3-D Object Recognition by a Service Robot

This paper presents a novel 3-D object recognition framework for a service robot to eliminate false detections in cluttered office environments where objects are in a great diversity of shapes and difficult to be represented by exact models. Laser point clouds are first converted to bearing angle images and a Gentleboost-based approach is then deployed for multiclass object detection. In order to solve the problem of variable object scales in object detection, a scale coordination technique is adopted in every subscene that is segmented from the whole scene according to the spatial distribution of 3-D laser points. Moreover, semantic information (e.g., ceilings, floors, and walls) extracted from raw 3-D laser points is utilized to eliminate false object detection results. K-means clustering and Mahalanobis distance are finally deployed to perform object segmentation in a 3-D laser point cloud accurately. Experiments were conducted on a real mobile robot to show the validity and performance of the proposed method.

A Robust Edge-Based Corner Detector (EBCD)

This paper presents a novel and robust edge-based corner detector (EBCD) which finds corners that are considered stable interest points in the framework of 2D object recognition especially for robot navigation. Using the EBCD, the corners are defined as intersection points of non-collinear straight image edges. The straight edge detector plus the corner detector will localize the corner positions. The detected corners have special features which are their angles and their sides length ratios. These features are invariant parameters which make the corners perfect for 2D object recognition. In addition, these corners are shown to be very robust against various image transformations like image scaling, rotation, translation and viewpoint illumination changes. Some updates are applied on the linking edge step in order to extract edges and their intersections that in turn construct the searched corners. Experiments conducted on synthetic and real images show that the proposed EBCD is able to achieve a very good performance in terms of accuracy, stability and especially computational efficiency in comparison with existing algorithms on interest points.

English

This paper presents a saliency based video object extraction and recognition framework. The extraction framework automatically extract foreground object of interest without any use of training data. The recognition system uses list of training data beforehand. For extracting the foreground object from a video, it uses visual and motion saliency features .A conditional random field is used to effectively combine the saliency induces features. Our proposed method is able to preserve spatial continuity and temporal consistency. Experiments results on variety video shows that our proposed system provides qualitatively satisfactory video object extraction (VOE) results.

Smart scanning and near real-time 3D surface modeling of dynamic construction equipment from a point cloud

Abstract This paper introduces a framework of automatic object recognition and rapid surface modeling to aid the heavy equipment operation in rapidly perceiving 3D working environment at dynamic construction sites. A custom-designed data acquisition system was employed in this study to rapidly recognize the selected target objects in a 3D space by dynamically separating target object’s point cloud data from a background scene for a quick computing process. A smart scanning method was also applied to only update the target object’s point cloud data while keeping the previously scanned static work environments. Then the target’s point cloud data were rapidly converted into a 3D surface model using the concave hull surface modeling algorithm after a process of data filtering and downsizing to increase the model accuracy and data processing speed. The performance of the proposed framework was tested with two different heavy equipment types at a steel frame building construction site. The generated surface model and the point cloud of static surroundings were wirelessly presented to a remote operator. The field test results show that the proposed rapid target surface modeling method would significantly improve productivity and safety in heavy construction equipment operations by distinguishing a dynamic target object from a surrounding static environment in 3D views in near real time.

Automatic segmentation of granular objects in images: Combining local density clustering and gradient-barrier watershed

Blob or granular object recognition is an image processing task with a rich application background, ranging from cell/nuclei segmentation in biology to nanoparticle recognition in physics. In this study, we establish a new and comprehensive framework for granular object recognition. Local density clustering and connected component analysis constitute the first stage. To separate overlapping objects, we further propose a modified watershed approach called the gradient-barrier watershed, which better incorporates intensity gradient information into the geometrical watershed framework. We also revise the marker-finding procedure to incorporate a clustering step on all the markers initially found, potentially grouping multiple markers within the same object. The gradient-barrier watershed is then conducted based on those markers, and the intensity gradient in the image directly guides the water flow during the flooding process. We also propose an important scheme for edge detection and fore/background separation called the intensity moment approach. Experimental results for a wide variety of objects in different disciplines – including cell/nuclei images, biological colony images, and nanoparticle images – demonstrate the effectiveness of the proposed framework.

Fisher tensors for classifying human epithelial cells

Analyzing and classifying Human Epithelial type 2 (HEp-2) cells using Indirect Immunofluorescence protocol has been the golden standard for detecting connective tissue diseases such as Rheumatoid Arthritis. However, this suffers from numerous shortcomings such as being subjective as well as time and labor intensive. Recently, several studies explore the advantages of artificial systems to automate the process, not only to reduce the test turn-around time but also to deliver more consistent results. In this paper, we extend the conventional bag of word models from Euclidean space to non-Euclidean Riemannian manifolds and utilize them to classify the HEp-2 cells. The main motivation comes from the observation that HEp-2 cells can be efficiently described by symmetric positive definite matrices which lie on a Riemannian manifold. With this motivation, we first discuss an intrinsic bag of Riemannian words model. We then propose Fisher tensors which can in turn encode additional information about the distribution of the signatures in a bag of word model. Experiments on two challenging HEp-2 images datasets, namely ICPRContest and SNPHEp-2 show that the proposed methods obtain notable improvements in discrimination accuracy, in comparison to baseline and several state-of-the-art methods. The proposed framework, while hand-crafted towards cell classification, is a generic framework for object recognition. This is supported by assessing the performance of our proposal on a challenging texture classification task.