Machine Vision Application on Science and Industry

Machine vision studies opens a great opportunity for different domains as manufacturing, agriculture, aquaculture, medical research, also research studies and applications for better understanding of processes and operations. As scientists' efforts had been directed towards deep understanding of the particular material systems or particular classes of types of specific fruits, or diagnosis of patients through medical images classification and analysis, also real time detection and inspection of malfunction piece, or process, as various domains witnessed advancement through using machine vision techniques and methods.

Application of Computer Vision Technology to Structural Health Monitoring of Engineering Structures

The computer vision technology has gained great advances and applied in a variety of industry fields. It has some unique advantages over the traditional technologies such as high speed, high accuracy, low noise, anti-electromagnetic interference, etc. In the last decade, the technology of computer vision has been widely employed in the field of structure health monitoring (SHM). Many specific hardware and algorithms have been developed to meet different kinds of monitoring demands. This chapter presents three application scenarios of computer vision technology for health monitoring of engineering structures, including bridge inspection and evaluation with unmanned aerial vehicle (UAV), recognition and surveillance of foreign object intrusion for railway system, and identification and tracking of concrete cracking. The principles and procedures of three application scenarios are addressed following with the experimental study, and the possibilities and ideas for the application of computer vision technology to other monitoring items are also discussed.

Object-Oriented Logic Programming of Intelligent Visual Surveillance for Human Anomalous Behavior Detection

The idea of the logic programming-based approach to the intelligent visual surveillance is in usage of logical rules for description and analysis of people behavior. New prospects in logic programming of the intelligent visual surveillance are connected with the usage of 3D machine vision methods and adaptation of the multi-agent approach to the intelligent visual surveillance. The main advantage of usage of 3D vision instead of the conventional 2D vision is that the first one can provide essentially more complete information about the video scene. The availability of exact information about the coordinates of the parts of the body and scene geometry provided by means of 3D vision is a key to the automation of behavior analysis, recognition, and understanding. This chapter supplies the first systematic and complete description of the method of object-oriented logic programming of the intelligent visual surveillance, special software implementing this method, and new trends in the research area linked with the usage of novel 3D data acquisition equipment.

Zeolite-Based Optical Detectors

The detection of chemical species is a common and required task in several areas of technology. Currently, measurements in dedicated labs are the predominant tools for detection and characterization of chemicals and materials. Although these techniques are available in specialized equipment, their use is often bounded by cost of application or the operator's expertise. Also, in many applications rather than an analysis of all the detectable chemical species, it is only of interest to determine the presence of a particular chemical compound, and if it is present, to quantify its concentration. For these reasons, alternative methods for detecting specific chemical species are required. One case of such methods are the optical chemical sensors, particularly the ones based on the materials known as zeolites. In a broad sense, these sensors are constituted by an optical detector that is modified with zeolites. This combination allows the detection of specific chemical compounds if the zeolitic materials is properly modified to have an optical response for the analyte.

Methods and Algorithms for Technical Vision in Radar Introscopy

Optimization of technical characteristics of radio vision systems is considered in the radars with ultra-wideband sounding signals. Highly noisy conditions, in which such systems operate, determine the requirements that should be met by the signals being studied. The presence of the multiplicative noise makes it difficult to design optimal algorithms of echo-signal processing. Consideration is being given to the problem of discriminating objects hidden under upper layers of the ground at depths comparable to the probing pulse duration. Based upon the cepstrum and textural analysis, a subsurface radar signal processing technique has been suggested. It is shown that, however the shape of the probing signal spectrum might be, the responses from point targets in the cepstrum images of subsurface ground layers make up the texture whose distinctive features enable objects to be detected and identified.

Moving Object Classification Under Illumination Changes Using Binary Descriptors

Object recognition and classification has become important in a surveillance video situated at prominent areas such as airports, banks, military installations, etc. Outdoor environments are more challenging for moving object classification because of incomplete appearance details of moving objects due to illumination changes and large distance between the camera and moving objects. As such, there is a need to monitor and classify the moving objects by considering the challenges of video in the real time. Training the classifiers using feature-based approaches is easier and faster than pixel-based approaches in object classification. Extraction of a set of features from the object of interest is most important for classification. Viewpoint and sources of light illumination plays major role in the appearance of an object. Abrupt transitions are identified using Chi-square and corners are detected using Harris corner detection. Silhouettes are captured using background subtraction and feature extraction is done using ORB. k-NN classifier is used for classification.

Mobile Robot Path Planning Using Continuous Laser Scanning

The main object of this book chapter is an introduction and presentation of mobile robot path planning using continuous laser scanning, which has significant advantages compared with discrete laser scanning. A general introduction to laser scanning systems is given, whereby a novel technical vision system (TVS) using the dynamic triangulation measurement method for 3D coordinate determination is found suitable for accomplishing this task of mobile robot path planning. Furthermore, methods and algorithms for mobile robot road maps and path planning are presented and compared.

Methods to Reduce the Optical Noise in a Real-World Environment of an Optical Scanning System for Structural Health Monitoring

This chapter describes different methods and devices that can be used in optical scanning systems (OSS), especially applied to structural health and monitoring (SHM) in order to reduce the interference and losing of resolution in the measurements of the displacements and coordinates calculated by the OSS of a specific structure to be monitored. The principal parts of the OSS are a photo-detector, non-rotating emitter source of light, a DC electrical motor, lens, and mirror. All the measurements and experiments have been realized in a controlled environmental; the optical noise was simulated with a similar intensity than the intensity of the reference signal of the emitter source. Applying analogue filters has disadvantages because part of signal with important information for the performance of the system is removed, but particularly the components will often be too costly. However, there are digital filters and techniques of computational statistics that can solve these problems.

Machine Vision-Based Application to Structural Analysis in Seismic Testing by Shaking Table

In the present chapter the most recent and successful experiences in the application of machine vision-based techniques to structural analysis with main focus on seismic testing by shaking table are described and discussed. In particular, the potentialities provided by 3D motion capture methodologies and, more recently, by motion magnification analysis (MMA) emerged as interesting integrations, if not alternatives, to more conventional and consolidated measurement systems in this field. Some examples of laboratory applications are illustrated with the aim of providing evidence and details on the practical potentialities and limits of these methodologies for vibration motion acquisition, as well as on data processing and analysis.

Leveraging Models of Human Reasoning to Identify EEG Electrodes in Images With Neural Networks

Humans have very little trouble recognizing discrete objects within a scene, but performing the same tasks using classical computer vision techniques can be counterintuitive. Humans, equipped with a visual cortex, perform much of this work below the level of consciousness, and by the time a human is conscious of a visual stimulus, the signal has already been processed by lower order brain regions and segmented into semantic regions. Convolutional neural networks are modeled loosely on the structure of the human visual cortex and when trained with data produced by human actors are capable of emulating its performance. By black-boxing the low-level image analysis tasks in this way, the authors model solutions to problems in terms of the workflows of expert human operators, leveraging both the work performed pre-consciously and the higher-order algorithmic solutions employed to solve problems.