Practical computer vision: Example techniques and challenges

Pankanti, Sharath; Brown, Lisa M.; Connell, Jonathan H.; Datta, Amit; Fan, Quanfu; Feris, Rogério; Haas, Norman; Li, Y.; Ratha, Nalini K.; Trinh, Hoang

doi:10.1147/jrd.2011.2165676

Cited by 8 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the real-time level, the proliferation of CCTV systems for public safety and their decreasing cost have increased the number of cameras deployed and operated on behalf of government authorities: in Dublin alone, there are approximately 700 CCTV cameras for public safety, making it very difficult to process all generated video feeds in real-time. One option is using machine vision on the video streams, and there are indeed effective solutions for this [21]. Nevertheless, they do not cover the situation where the cause of the change is not immediately visible.…”

Section: Operationalmentioning

confidence: 99%

Real-Time Urban Monitoring in Dublin Using Semantic and Stream Technologies

Tallevi-Diotallevi

Kotoulas

Foschini

et al. 2013

Advanced Information Systems Engineering

View full text Add to dashboard Cite

Several sources of information, from people, systems, things, are already available in most modern cities. Processing these continuous flows of information and capturing insight poses unique technical challenges that span from response time constraints to data heterogeneity, in terms of format and throughput. To tackle these problems, we focus on a novel prototype to ease realtime monitoring and decision-making processes for the City of Dublin with three main original technical aspects: (i) an extension to SPARQL to support efficient querying of heterogeneous streams; (ii) a query execution framework and runtime environment based on IBM InfoSphere Streams, a high-performance, industrial strength, stream processing engine; (iii) a hybrid RDFS reasoner, optimized for our stream processing execution framework. Our approach has been validated with real data collected on the field, as shown in our Dublin City video demonstration. Results indicate that real-time processing of city information streams based on semantic technologies is indeed not only possible, but also efficient, scalable and low-latency.

show abstract

Section: Operationalmentioning

confidence: 99%

Real-Time Urban Monitoring in Dublin Using Semantic and Stream Technologies

Tallevi-Diotallevi

Kotoulas

Foschini

et al. 2013

Advanced Information Systems Engineering

View full text Add to dashboard Cite

show abstract

“…To accurately detect vehicles, we should effectively address three major challenges: location ambiguity, size ambiguity, and appearance variation of vehicles [1], [2]. These challenges have been overcome by sliding window based detection methods [2]- [5] because of their superior performance [6], [7] compared with other approaches employing branch-andbound search [8], [9] and coarse-to-fine search [10]- [12].…”

Section: Introductionmentioning

confidence: 99%

Vehicle Detection Using Local Size-Specific Classifiers

Noh

Jeon

2016

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYAs the number of surveillance cameras keeps increasing, the demand for automated traffic-monitoring systems is growing. In this paper, we propose a practical vehicle detection method for such systems. In the last decade, vehicle detection mainly has been performed by employing an image scan strategy based on sliding windows whereby a pre-trained appearance model is applied to all image areas. In this approach, because the appearance models are built from vehicle sample images, the normalization of the scales and aspect ratios of samples can significantly influence the performance of vehicle detection. Thus, to successfully apply sliding window schemes to detection, it is crucial to select the normalization sizes very carefully in a wise manner. To address this, we present a novel vehicle detection technique. In contrast to conventional methods that determine the normalization sizes without considering given scene conditions, our technique first learns local region-specific size models based on scenecontextual clues, and then utilizes the obtained size models to normalize samples to construct more elaborate appearance models, namely local sizespecific classifiers (LSCs). LSCs can provide advantages in terms of both accuracy and operational speed because they ignore unnecessary information on vehicles that are observable in faraway areas from each sliding window position. We conduct experiments on real highway traffic videos, and demonstrate that the proposed method achieves a 16% increased detection accuracy with at least 3 times faster operational speed compared with the state-of-the-art technique.

show abstract

“…Recently, many machine-vision based automatic systems have been introduced to solve different tasks in rail inspection [10,7,1,2,3,4,5,9], including our own work [6,8]. Each of these systems is focused on a different subset of rail inspection problems.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Enhanced rail component detection and consolidation for rail track inspection

Trinh

Haas

Liu

et al. 2012

2012 IEEE Workshop on the Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

For safety purposes, railroad tracks need to be inspected on a regular basis for physical defects or design noncompliances. Such track defects and non-compliances, if not detected in a timely manner, may eventually lead to grave consequences such as train derailments.In this paper, we present a real-time automatic visionbased rail inspection system, with main focus on anchorsan important rail component type, and anchor-related rail defects, or exceptions. Our system robustly detects important rail components including ties, tie plates, anchors with high accuracy and efficiency. Detected objects are then consolidated across video frames and across camera views to map to physical rail objects, by combining the video data streams from all camera views with GPS information and speed information from the distance measuring instrument (DMI). After these rail components are detected and consolidated, further data integration and analysis is followed to detect sequence-level track defects, or exceptions. Quantitative analysis performed on a real online field test conducted on different track conditions demonstrates that our system achieves very promising performance in terms of rail component detection, anchor condition assessment, and compliance-level exception detection. We also show that our system outperforms another advanced rail inspection system in anchor detection.

show abstract

Practical computer vision: Example techniques and challenges

Cited by 8 publications

References 23 publications

Real-Time Urban Monitoring in Dublin Using Semantic and Stream Technologies

Real-Time Urban Monitoring in Dublin Using Semantic and Stream Technologies

Vehicle Detection Using Local Size-Specific Classifiers

Enhanced rail component detection and consolidation for rail track inspection

Contact Info

Product

Resources

About