Antonios Gasteratos

Review of stereo vision algorithms: From software to hardware

Stereo vision, resulting in the knowledge of deep information in a scene, is of great importance in the field of machine vision, robotics and image analysis. In this article, an explicit analysis of the existing stereo matching methods, up to date, is presented. The presented algorithms are discussed in terms of speed, accuracy, coverage, time consumption, and disparity range. Towards the direction of real-time operation, the development of stereo matching algorithms, suitable for efficient hardware implementation is highly desirable. Implementations of stereo matching algorithms in hardware for real-time applications are also discussed in details.

Semantic mapping for mobile robotics tasks

The evolution of contemporary mobile robotics has given thrust to a series of additional conjunct technologies. Of such is the semantic mapping, which provides an abstraction of space and a means for human-robot communication. The recent introduction and evolution of semantic mapping motivated this survey, in which an explicit analysis of the existing methods is sought. The several algorithms are categorized according to their primary characteristics, namely scalability, inference model, temporal coherence and topological map usage. The applications involving semantic maps are also outlined in the work at hand, emphasizing on human interaction, knowledge representation and planning. The existence of publicly available validation datasets and benchmarking, suitable for the evaluation of semantic mapping techniques is also discussed in detail. Last, an attempt to address open issues and questions is also made. Two level navigation.Cognitive navigation.Spatial semantics.

Disparity Estimation on Log-Polar Images and Vergence Control

An important issue in the realization of an autonomous robot with stereoscopic vision is the control of vergence. Together with version, it determines uniquely the position of the fixation point in space. Vergence control is directly related to both depth perception and binocular fusion. Previous works in this field employed either a measure of correlation of stereo images or some kind of disparity-related estimate. In this paper, we present a new method of extracting a global disparity measure for vergence control, which does not require a priori segmentation of the object of interest. Our method uses images acquired by retina-like sensors and, therefore, the computation is performed in the log-polar plane. The technique we present here is: (i) global, in the sense that it is an integral measure over the whole image, (ii) computationally inexpensive, considering that the goal was to use it in the robot control loop rather than to accurately measure some 3D world features. Moreover, the proposed technique is robust and independent of the average illumination as well as of other features of the target such as size, shape, and direction of motion. It provides a precise and linear estimate of the vergence error, which is the only requirement from the control point of view. Several experimental results on a real robotic setup demonstrate the effectiveness of the proposed technique.

Image moment invariants as local features for content based image retrieval using the Bag-of-Visual-Words model

A new image descriptor specifically designed for image retrieval tasks is introduced.Evaluation of affine moment invariants in the area of image retrieval.The usage of image chromaticities improves the overall retrieval performance. This paper presents an image retrieval framework that uses affine image moment invariants as descriptors of local image areas. Detailed feature vectors are generated by feeding the produced moments into a Bag-of-Visual-Words representation. Image moment invariants have been selected for their compact representation of image areas as well as due to their ability to remain unchanged under affine image transformations. Three different setups were examined in order to evaluate and discuss the overall approach. The retrieval results are promising compared with other widely used local descriptors, allowing the proposed framework to serve as a reference point for future image moment local descriptors applied to the general task of content based image retrieval.

Learning spatially semantic representations for cognitive robot navigation

Contemporary mobile robots should exhibit enhanced capacities, which allow them self-localization and semantic interpretation as they move into an unexplored environment. The coexistence of accurate SLAM and place recognition can provide a descriptive and adaptable navigation model. In this paper such a two-layer navigation scheme is introduced suitable for indoor environments. The low layer comprises a 3D SLAM system based solely on an RGB-D sensor, whilst the high one employs a novel content-based representation algorithm, suitable for spatial abstraction. In course of robots locomotion, salient visual features are detected and they shape a bag-of-features problem, quantized by a Neural Gas to code the spatial information for each scene. The learning procedure is performed by an SVM classifier able to accurately recognize multiple dissimilar places. The two layers mutually interact with a semantically annotated topological graph augmenting the cognition attributes of the integrated system. The proposed framework is assessed on several datasets, exhibiting remarkable accuracy. Moreover, the appearance based algorithm produces semantic inferences suitable for labeling unexplored environments.

Biologically and psychophysically inspired adaptive support weights algorithm for stereo correspondence

In this paper a novel stereo correspondence algorithm is presented. It incorporates many biologically and psychologically inspired features to an adaptive weighted sum of absolute differences (SAD) framework in order to determine the correct depth of a scene. In addition to ideas already exploited, such as the color information utilization, gestalt laws of proximity and similarity, new ones have been adopted. The presented algorithm introduces the use of circular support regions, the gestalt law of continuity as well as the psychophysically-based logarithmic response law. All the aforementioned perceptual tools act complementarily inside a straightforward computational algorithm applicable to robotic applications. The results of the algorithm have been evaluated and compared to those of similar algorithms.

Fault diagnosis of photovoltaic modules through image processing and Canny edge detection on field thermographic measurements

Today, conventional condition monitoring of installed, operating photovoltaic (PV) modules is mainly based on electrical measurements and performance evaluation. However, such practices exhibit restricted fault-detection ability. This study proposes the use of standard thermal image processing and the Canny edge detection operator as diagnostic tools for module-related faults that lead to hot-spot heating effects. The intended techniques were applied on thermal images of defective PV modules, from several field infrared thermographic measurements conducted during this study. The whole approach provided promising results with the detection of hot-spot formations that were diagnosed to specific defective cells in each inspected module. These evolving hot spots lead to abnormally low performance of the PV modules, a fact that is also validated by the manufacturers standard electrical tests.

Stereovision-Based Algorithm for Obstacle Avoidance

This work presents a vision-based obstacle avoidance algorithm for autonomous mobile robots. It provides an efficient solution that uses a minimum of sensors and avoids, as much as possible, computationally complex processes. The only sensor required is a stereo camera. The proposed algorithm consists of two building blocks. The first one is a stereo algorithm, able to provide reliable depth maps of the scenery in frame rates suitable for a robot to move autonomously. The second building block is a decision making algorithm that analyzes the depth maps and deduces the most appropriate direction for the robot to avoid any existing obstacles. The proposed methodology has been tested on sequences of self-captured outdoor images and its results have been evaluated. The performance of the algorithm is presented and discussed.

A Dense Stereo Correspondence Algorithm for Hardware Implementation with Enhanced Disparity Selection

In this paper an effective, hardware oriented stereo correspondence algorithm, able to produce dense disparity maps of improved fidelity is presented. The proposed algorithm combines rapid execution, simple and straight-forward structure as well as comparably high quality of results. These features render it as an ideal candidate for hardware implementation and for real-time applications. The proposed algorithm utilizes the Absolute Differences (AD) as matching cost and aggregates the results inside support windows, assigning Gaussian distributed weights to the support pixels, based on their Euclidean distance. The resulting Disparity Space Image (DSI) is furthered refined by Cellular Automata (CA) acting in all of the three dimensions of the DSI. The algorithm is applied to typical as well as to self-recorded real-life image sets. The disparity maps obtained are presented and quantitatively examined.

REALIZATION OF RANK ORDER FILTERS BASED ON MAJORITY GATE

A new technique for the implementation of a single hardware structure capable of computing any rank order filter is presented in this paper. The proposed technique, which is based on the majority gate, achieves faster extraction of setting flag signals and, therefore, shorter processing times are attained. A pipelined systolic array, suitable for performing rank order filtering, is also presented. Applications of rank order filters include digital image processing, speech processing and coding and digital TV applications.