Nicholas Miller, Mohan A. Thomas, Justin A. Eichel, A. Mishra
{"title":"A Hidden Markov Model for Vehicle Detection and Counting","authors":"Nicholas Miller, Mohan A. Thomas, Justin A. Eichel, A. Mishra","doi":"10.1109/CRV.2015.42","DOIUrl":"https://doi.org/10.1109/CRV.2015.42","url":null,"abstract":"To reduce roadway congestion and improve traffic safety, accurate traffic metrics, such as number of vehicles travelling through lane-ways, are required. Unfortunately most existing infrastructure, such as loop-detectors and many video detectors, do not feasibly provide accurate vehicle counts. Consequently, a novel method is proposed which models vehicle motion using hidden Markov models (HMM). The proposed method represents a specified small region of the roadway as 'empty', 'vehicle entering', 'vehicle inside', and 'vehicle exiting', and then applies a modified Viterbi algorithm to the HMM sequential estimation framework to initialize and track vehicles. Vehicle observations are obtained using an Adaboost trained Haar-like feature detector. When tested on 88 hours of video, from three distinct locations, the proposed method proved to be robust to changes in lighting conditions, moving shadows, and camera motion, and consistently out-performed Multiple Target Tracking (MTT) and Virtual Detection Line(VDL) implementations. The median vehicle count error of the proposed method is lower than MTT and VDL by 28%, and 70% respectively. As future work, this algorithm will be implemented to provide the traffic industry with improved automated vehicle counting, with the intent to eventually provide real-time counts.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125913674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Reconstruction of 3-D Density Functions from Few Projections: Structural Assumptions for Graceful Degradation","authors":"M. Cormier, D. Lizotte, R. Mann","doi":"10.1109/CRV.2015.27","DOIUrl":"https://doi.org/10.1109/CRV.2015.27","url":null,"abstract":"We present a spatial-domain method for reconstructing a three-dimensional density distribution from one or more projections (images formed by integration of density along lines of sight) and using the three-dimensional reconstruction to explain features of the two-dimensional images. The advantages of our proposed method are that it degrades gracefully down to a single image, that it uses linear equations and constraints (allowing the use of convex optimization), that it is amenable to three-dimensional structural biases, and that ambiguity can be expressed precisely (it is possible to \"know what we don't know\"). Previously described methods have some, but not all, of these properties.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127009452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ayan Chaudhury, Christopher D. W. Ward, A. Talasaz, A. Ivanov, N. Hüner, B. Grodzinski, Rajnikant V. Patel, J. Barron
{"title":"Computer Vision Based Autonomous Robotic System for 3D Plant Growth Measurement","authors":"Ayan Chaudhury, Christopher D. W. Ward, A. Talasaz, A. Ivanov, N. Hüner, B. Grodzinski, Rajnikant V. Patel, J. Barron","doi":"10.1109/CRV.2015.45","DOIUrl":"https://doi.org/10.1109/CRV.2015.45","url":null,"abstract":"Research on increasing the production of crops is increasingly important these days. This research needs a way to quantitatively measure the 3D growth of plants under controlled environments to allow a cost versus benefits analysis. Plant scientists need a non-invasive, non-destructive method to quantitatively measure the 3D growth of plants. Traditional methods, for example, measuring weight, area or volume, often negatively affects the future plant growth. Also the manual nature of this measurement can be quite time consuming, tedious and error prone. Some recent effort have been reported in the literature about the construction of autonomous systems for plant phenotype, but these are not practical for large scale accurate 3D plant growth computation. To the best of our knowledge, we are the first in the world to attempt truly 3D approach via robot assisted plant growth analysis using 3D imaging and laser scanning technology. We describe an automated system to perform 3D plant modelling using a laser scanner mounted on a robot arm to capture 3D plant data. We present a detailed overview of the system integration, including the robotic arm, laser scanner and a programmable growth chamber. We also show some results on reconstructing the 3Dmodel of a growing plant which is better than the current state of the art.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115885208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Shrink Wrapping Small Objects","authors":"Sricharana Rajagopal, Kaleem Siddiqi","doi":"10.1109/CRV.2015.44","DOIUrl":"https://doi.org/10.1109/CRV.2015.44","url":null,"abstract":"We consider the problem of scanning and modelling small objects using easily available RGBD sensors such as the Kinect. These sensors allow for the acquisition of view-based registered depth and color data. We register the point clouds by recovering the rigid transformation between successive pairs in a sequence of views, and then demonstrate the utility of surface evolution for shrink wrapping the result using geometric flow based approaches for surface modelling. In our experiments we use a PrimeSense Carmine 1.09, which is a high resolution short range RGBD sensor suited to capture small objects. We obtain high quality surface and appearance models of small objects that are competitive with commercial software packages while better capturing surface detail.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115901721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Nakhaeinia, P. Laferriere, P. Payeur, R. Laganière
{"title":"Safe Close-Proximity and Physical Human-Robot Interaction Using Industrial Robots","authors":"D. Nakhaeinia, P. Laferriere, P. Payeur, R. Laganière","doi":"10.1109/CRV.2015.38","DOIUrl":"https://doi.org/10.1109/CRV.2015.38","url":null,"abstract":"Industrial robots have been employed worldwide in the manufacturing sector for performing tasks quickly, repeatedly and accurately in relatively static environments for over 30 years. In recent years, close physical interaction between industrial robots and human operators has attracted researchers' attention and encouraged a number of technological innovations to turn these robots into human-robot platforms. In this work a specially designed compliant wrist is developed to support dexterous robotic interaction with live proximity and contact feedback. The compliant wrist incorporates a level of compliance into an initially non-compliant manipulator robot which allows the robot to dynamically adapt to the surfaces it approaches or touches. Furthermore, to facilitate human -- robot interactions, the robot must be able to adapt its behaviour to the human partner. Therefore, a real-time path planning method is developed to generate online motion, adapt the robot to dynamic changes in the environment and ensure smooth interactions. The performance of the proposed method is demonstrated through experimental results on a CRS-F3 manipulator.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"307 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121742121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Being in Two Places at Once: Smooth Visual Path Following on Globally Inconsistent Pose Graphs","authors":"Sebastian Kai van Es, T. Barfoot","doi":"10.1109/CRV.2015.17","DOIUrl":"https://doi.org/10.1109/CRV.2015.17","url":null,"abstract":"Early work in the field of SLAM asserted that globally metrically consistent maps expressed in a single coordinate frame were necessary for autonomous operation. It has been shown previously that chain-structured and tree-structured optometric maps provide sufficient information for accurate path following. This paper extends this concept to arbitrarily connected graph structures with loop closures. We show that globally inconsistent maps may be treated as a set of locally defined Riemannian manifolds, and that this representation is sufficient for path repetition tasks. We demonstrate smooth path following on an inconsistent optometric map with loop closures, using the existing Visual Teach and Repeat (VT&R) framework for vision-in-the-loop control. Path-tracking errors are maintained within nominal values despite disparities of over 2m between the local and global representations of robot pose. Traversal of large map discontinuities is found to have no adverse effect on robot performance, allowing segments of the map to be repeated in a different order than they were trained.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129974918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of a Low Cost Gamma-Ray Imaging System Using Handheld Scintillation Detectors for Visual Surveying of Radiation Fields with Robots","authors":"A. Miller, R. Machrafi","doi":"10.1109/CRV.2015.14","DOIUrl":"https://doi.org/10.1109/CRV.2015.14","url":null,"abstract":"This paper presents a novel method for making an inexpensive, low resolution, wide angle gamma-ray camera using a single handheld inorganic scintillation detector. A lead aperture is used to create a directionally sensitive detector and craterisation points are measured and compiled into a pixelated spherical gamma-ray image. The gamma-ray image is separated into energy channels and overlaid on top of a grey scale visual panoramic image so that collared pixels indicate the position and energy of gamma-ray sources.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115543472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Uncertainty Reduction via Heuristic Search Planning on Hybrid Metric/Topological Map","authors":"Qiwen Zhang, Ioannis M. Rekleitis, G. Dudek","doi":"10.1109/CRV.2015.36","DOIUrl":"https://doi.org/10.1109/CRV.2015.36","url":null,"abstract":"This paper presents an extension of our previous work on hybrid metric/topological maps to enable uncertainty reduction planning through the map, taking into account both map uncertainty and distance. An enhancement of the edge structure which enables the simulation of bidirectional edge propagation through an extended Kalman filter is proposed in our heuristic search planning algorithm to plan for maximal map uncertainty reduction. This work expands on the heuristic search framework proposed in [1] to apply in hybrid metric/topological maps instead of more constrained camera sensor networks. Experimental results from realistic simulations and deployment on a real robotic system are presented to show the efficacy of the proposed algorithm and validate our approach for uncertainty reduction.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115718405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Feature Ranking in Dynamic Texture Clustering","authors":"T. Nguyen, Q.M. Jonathan Wu, D. Mukherjee","doi":"10.1109/CRV.2015.23","DOIUrl":"https://doi.org/10.1109/CRV.2015.23","url":null,"abstract":"The dynamic texture clustering over space and time, which considers a video as a sample from a linear dynamical system, has recently received great attention. While there exist many algorithms for time-varying characteristics and phenomena clustering, an important and challenging issue arising from these studies concerns is that how to compare the importance of each feature, and which features of the data should be used for time-varying clustering. In view of the aforementioned observations, a new unsupervised feature ranking for dynamic texture clustering is proposed in this paper. Different from previous works, where the selection criteria are often used to identify relevant features, our method has an ability to score relevant features, and allow us to evaluate the importance of each feature. Also, it is intuitively appealing, avoiding any combinatorial search. Numerical experiments on various time-varying characteristics and phenomena demonstrate the accuracy and effectiveness of the proposed method.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126758941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An Online Unsupervised Feature Selection and its Application for Background Suppression","authors":"T. Nguyen, Q. M. J. Wu, D. Mukherjee","doi":"10.1109/CRV.2015.53","DOIUrl":"https://doi.org/10.1109/CRV.2015.53","url":null,"abstract":"Background suppression in video sequences has recently received great attention. While there exist many algorithms for background suppression, an important and challenging issue arising from these studies concerns that for which attributes of the data should be used for background modelling. It is interesting and difficult because there is no knowledge about the data to guide the search. Also, in real application for background suppression, the video length is unknown and the video frames are generated dynamically in a streaming fashion and arrive one at a time. Thus, it is impractical to wait until all data have been generated before feature learning begins. In this paper, we present an online unsupervised feature selection for background suppression. The advantage of our method is that it avoids any combinatorial search, is intuitively appealing, and allows us to prune the feature set. Moreover, our method, based on the self-adaptive model, has an ability to adapt and change through complex scenes. Experiments on real-world datasets are conducted. The performance of the proposed model is compared to that of other background modelling techniques, demonstrating the robustness and accuracy of our method.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124070818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}