Machine Vision Approaches Research Articles

A rule-based event detection system for real-life underwater domain

Understanding and analyzing fish behaviour is a fundamental task for biologists that study marine ecosystems because the changes in animal behaviour reflect environmental conditions such as pollution and climate change. To support investigators in addressing these complex questions, underwater cameras have been recently used. They can continuously monitor marine life while having almost no influence on the environment under observation, which is not the case with observations made by divers for instance. However, the huge quantity of recorded data make the manual video analysis practically impossible. Thus machine vision approaches are needed to distill the information to be investigated. In this paper, we propose an automatic event detection system able to identify solitary and pairing behaviours of the most common fish species of the Taiwanese coral reef. More specifically, the proposed system employs robust low-level processing modules for fish detection, tracking and recognition that extract the raw data used in the event detection process. Then each fish trajectory is modeled and classified using hidden Markov models. The events of interest are detected by integrating end-user rules, specified through an ad hoc user interface, and the analysis of fish trajectories. The system was tested on 499 events of interest, divided into solitary and pairing events for each fish species. It achieved an average accuracy of 0.105, expressed in terms of normalized detection cost. The obtained results are promising, especially given the difficulties occurring in underwater environments. And moreover, it allows marine biologists to speed up the behaviour analysis process, and to reliably carry on their investigations.

A Driver Face Monitoring System for Fatigue and Distraction Detection

Driver face monitoring system is a real-time system that can detect driver fatigue and distraction using machine vision approaches. In this paper, a new approach is introduced for driver hypovigilance (fatigue and distraction) detection based on the symptoms related to face and eye regions. In this method, face template matching and horizontal projection of top-half segment of face image are used to extract hypovigilance symptoms from face and eye, respectively. Head rotation is a symptom to detect distraction that is extracted from face region. The extracted symptoms from eye region are (1) percentage of eye closure, (2) eyelid distance changes with respect to the normal eyelid distance, and (3) eye closure rate. The first and second symptoms related to eye region are used for fatigue detection; the last one is used for distraction detection. In the proposed system, a fuzzy expert system combines the symptoms to estimate level of driver hypo-vigilance. There are three main contributions in the introduced method: (1) simple and efficient head rotation detection based on face template matching, (2) adaptive symptom extraction from eye region without explicit eye detection, and (3) normalizing and personalizing the extracted symptoms using a short training phase. These three contributions lead to develop an adaptive driver eye/face monitoring. Experiments show that the proposed system is relatively efficient for estimating the driver fatigue and distraction.

Machine Vision Systems for Plant Growth Measurement and Modeling

Recent developments of microcomputer-based machine vision systems has offered convenient and non-destructive ways for measurements of plant characteristics that allow plant growth assessment. This paper presents brief reviews and comparisons of machine vision approaches for non-destructive plant growth measurement. The simple approach of using projected silhouette image of a plant was most commonly used in acquiring growth data in various experiments. Plant fresh or dry weight can be indirectly estimated from projected leaf area by calibration with data from standard measurement methods. However, the estimation error usually increases as the extent of overlapped leaves increases. Images acquired from different views or utilizing dual cameras allow the estimation of leaf area without predetermined calibration relationship but the image processing algorithms usually become more sophisticated. By incorporating multiple images of a plant, three-dimensional and structural information may be extracted for more detailed growth analyses and modeling.

Monitoring of Flotation Processes Using Multiresolutional Multivariate Image Analysis (MR-MIA)

This paper provides a novel machine vision solution based on multiresolutional multivariate image analysis (MR-MIA) for monitoring of flotation processes. The approach based on MR-MIA is superior to the contemporary machine vision approaches in terms of efficient analysis of morphological and color information and robustness to lighting conditions. The results show that the MR-MIA based approach provides enough morphological and color information of froth needed to describe different froth status. A PCA model built from the information provided by MR-MIA can provide multivariate control charts, from which the transition as well as steady-state process status can be monitored in real time.

FhG-Co-driver: From map-guided automatic driving by machine vision to a cooperative driver support

A digital road map provides partial knowledge about the operating environment for a road vehicle. If a road vehicle is equipped with a video camera, machine vision approaches can provide knowledge about the actual traffic environment around the vehicle. Experiences with a combination of two such approaches during the commissioning of a van for automatic driving on a private road network are reported, including experiences gathered during subsequent driving experiments on public roads and several improvement cycles for hardware and software. Based on these experiences, a second generation vehicle for automatic driving—a sedan—has been designed and commissioned. It is currently evaluated on public roads. This equipment provides an experimental platform for studying driver-vehicle interactions with the option to automatically evaluate actual traffic situations around the vehicle in real-time. Our equipment thus offers an approach to record and disentangle the multitude of factors which influence the—often subconscious—reactions of a driver. It is our working hypothesis that only an automatic, in-depth understanding of the actual traffic situation facilitates the design of a driver support system which is competent and flexible enough to win acceptance by a wide spectrum of users.

System Components for Driver Support by Evaluation of Image Sequences

Based on experiences by commissioning two experimental vehicles, a system architecture for investigating driver assistance approaches is described. If a road vehicle is equipped with a video camera, machine vision approaches can provide knowledge about the actual traffic situation around the vehicle, e.g. obstacles. Taking into consideration the complexity of traffic situations, a modelling approach for traffic scenes is necessary which is based on the instantiation of generic situation schemata within an interpretation cycle by utilizing sensor information about driver, vehicle and environment.

Shape analysis of complex profiles

This paper proposes a novel shape representation scheme based on CAD database specifications, and an optimization matching algorithm using the least-squares criterion to determine the pose and scale of objects, and perform out-of-profile inspection. The shapes of interest are two-dimensional (2D) profiles generated by projecting 3D objects onto a 2D inspection plane, the boundaries of which are composed of straight-line segments and circular arcs. This procedure can be applied to automated inspection of machined parts using machine vision systems. The shape model is characterized by four global parameters, based on which all the geometric primitives of the shape are expressed in analytic form. The matching procedure is to find the least-squares error between the true profile of the object and the sensed data, and to satisfy all the geometric constraints among primitives simultaneously. The advantage of the proposed model is that all the primitives of the shape are represented by the same (four) parameters in implicit equations, so that the spatial relationships among them are completely specified by the equations. Based on this representation scheme, the subsequent matching problem is formulated as an unconstrained optimization model, and solved by a gradient-based descent algorithm. In the experimental study, the results generated from the proposed technique are compared with those from two of the most commonly used machine vision approaches: Fourier Descriptors (FDs) and moments. This comparison shows that the method is superior to FDs and moments in error minimization. In addition, a complete procedure for automated design, manufacturing and machine vision inspection of production parts is described.

Machine-vision techniques for inspection of printed wiring boards and thick-film circuits

In this paper, a review of machine-vision methods for microelectronic production and related unsolved problems is given, based on research paradigms and manufacturing needs that are seldom addressed in the literature. Although the importance of many industrial machine-vision applications has been identified, this paper will cover only a subset of these problems because of space limitations. Specifically, the automated visual inspection of printed circuit boards and thick-film circuits will be reviewed. Different aspects of these problems will be surveyed, including manufacturing considerations, contributions to sensing techniques, and, more importantly, digital image processing and analysis methods. For printed wiring boards, several production stages are addressed in which automated visual inspection plays a key role, from artwork to lamination of multiple layers and plated-through holes. In addition, the main inspection problems arising in populated boards will be reviewed, such as solder joint integrity and top-surface component placement. Finally, some machine-vision approaches used for thick-film and hybrid circuit inspection are considered.