Browsing by Author "Rodrigo, BKRP"

Now showing 1 - 20 of 20

item: Conference-Abstract
Abnormal activity recognition using spatio-temporal features
Chathuramali, KGM; Ramasinghe, S; Rodrigo, BKRP
Abnormal activity detection plays an important role in many areas such as surveillance, military installations, and sports. Existing abnormal activity detectors mostly rely on motion data obtained over a number of frames to characterize abnormality. However, only motion may not be able to capture all forms of abnormality, in particular, poses that do not amount to motion “outliers”. In this paper, we propose two different spatiotemporal descriptors, a silhouette and optic flow based method and a dense trajectory based method which additionally include trajectory shape descriptor, to detect abnormalities. These two descriptors enable us to classify abnormal versus non-abnormal activities using SVM. Comparison with existing methods, using five standard datasets, shows that dense trajectory based method outperforms state-of-the-art results in crowd dataset and silhouette and optic flow based method outperforms others in some datasets.
item: Conference-Abstract
Action recognition by single stream convolutional neural networks : an approach using combined motion and static information
Ramasinghe, S; Rodrigo, BKRP
We investigate the problem of automatic action recognition and classification of videos. In this paper, we present a convolutional neural network architecture, which takes both motion and static information as inputs in a single stream. We show that the network is able to treat motion and static information as different feature maps and extract features off them, although stacked together. We trained and tested our network on Youtube dataset. Our network is able to surpass state-of-the-art hand-engineered feature methods. Furthermore, we also studied and compared the effect of providing static information to the network, in the task of action recognition. Our results justify the use of optic flows as the raw information of motion and also show the importance of static information, in the context of action recognition.
item: Conference-Abstract
Appearance based tracking with background subtraction
(2014-06-24) Jayamanne, DJ; Samarawickrama, J; Rodrigo, BKRP
Grouping the detected feature points traditionally requires the storage of long corner tracks. The traditional method does not permit to arrive at a decision to cluster the feature points based on a frame by frame basis. This paper presents a method to group the feature points directly into objects using the most recent 20 frames. The detected corner features are validated and clustered based on two approaches. When objects move in isolation, an EM algorithm is used to cluster and every object is detected and tracked. When objects move under partial occlusion, the corner features are clustered based on an agglomerative hierarchical clustering approach. A probabilistic framework has also been applied to determine the object level membership of the candidate corner features. A novel foreground estimation algorithm with an accuracy of 98% based on color information, background subtraction result and detected corner features is also presented.
item: Conference-Abstract
Automatic number plate recognition in low quality videos
(2014-06-20) Ajanthan, T; Kamalaruban, P; Rodrigo, BKRP
Typical Automatic Number Plate Recognition (ANPR) system uses high resolution cameras to acquire good quality images of the vehicles passing through. In these images, license plates are localized, characters are segmented, and recognized to determine the identity of the vehicles. However, the steps in this workflow will fail to produce expected results in low resolution images and in a less constrained environment. Thus in this work, several improvements are made to this ANPR workflow by incorporating intelligent heuristics, image processing techniques and domain knowledge to build an ANPR system that is capable of identifying vehicles even in low resolution video frames. Main advantages of our system are that it is able to operate in real-time, does not rely on special hardware, and not constrained by environmental conditions. Low quality surveillance video data acquired from a toll system is used to evaluate the performance of our system. We were able to obtain more than 90% plate level recognition accuracy. The experiments with this dataset have shown that the system is robust to variations in illumination, view point, and scale.
item: Conference-Abstract
Computationally efficient implementation of video rectification in an FPGA for stereo vision applications
(2016-08-29) Maldeniya, B; Nawarathna, D; Wijayasekara, K; Wijegoonasekara, T; Rodrigo, BKRP
Abstract—In order to obtain depth perception in computer vision, it is needed to process pairs of stereo images. This process is computationally challenging to be carried out in real-time, because it requires the search for matches between objects in both images. Such process is significantly simplified if the images are rectified. Stereo image rectification involves a matrix transformation which when done in software will not produce real-time results although it is very demanding. Therefore, the video streaming and matrix transformation are not usually implemented in the same system. Our product is a stereo camera pair which produces a rectified real time image output with a resolution of 320x240 at a frame rate of 15FPS and delivers them via a 100-Ethernet interface. We use a Spartan 3E FPGA for real-time processing within which we implement an image rectification algorithm.
item: Conference-Full-text
Curvature based robust descriptors
(2012) Mohideen, F; Rodrigo, BKRP
Feature descriptors have enabled feature matching under varying imaging conditions, while mostly being backed by experimental evidence. In addition to imposing some re- strictions in imaging conditions needed to ensure matching, extending the existing de- scriptors is not straightforward due to the lack of sound mathematical bases. In this work, by using a surface bending versus shape histogram based on the principal curvatures, we are able to produce a descriptor which is not sensitive to the errors in dominant orientation assignment. Experimental evaluations show that our descriptor outperforms existing descriptors in the areas of viewpoint, rotation, scale, zoom, lighting and compression changes, with the exception of resilience to blur. Further, we apply this descriptor for accuracy demanding applications such as homography estimation and pose estimation. The experimental results show significant improvements in estimated homography and pose in terms of residual error and Sampson distance respectively.
item: Conference-Abstract
Establishing object correspondence across non-overlapping calibrated cameras
Jayamanne, DJ; Rodrigo, BKRP
When establishing object correspondence across non-overlapping cameras, the existing methods combine separate likelihoods of appearance and kinematic features in a Bayesian framework, constructing a joint likelihood to compute the probability of re-detection. A drawback of these methods is not having a proper approach to reduce the search space when localizing an object in a subsequent camera once the kinematic and appearance features are extracted in the current camera. In this work we introduce a novel methodology to condition the location of an object on its appearance and time, without assuming independence between appearance and kinematic features, in contrast to existing work. We characterize the linear movement of objects in the unobserved region with an additive Gaussian noise model. Assuming that the cameras are affine, we transform the noise model onto the image plane of subsequent cameras. We have tested our method with toy car experiments and real-world camera setups and found that the proposed noise model acts as a prior to improving re-detection. It constrains the search space in a subsequent camera, greatly improving the computational efficiency. Our method also has the potential to distinguish between objects similar in appearance, and recover correct labels when they move across cameras.
item: Conference-Abstract
Extensible video surveillance software with simultaneous event detection for low and high density crowd analysis
Hettiarachchi, AL; Thathsarani, HO; Wickramasinghe, PU; Wickramasuriya, DS; Rodrigo, BKRP
Manual analysis of large volumes of video surveillance footage stemming from the widespread deployment of security cameras is error prone, expensive and time consuming. Despite the commercial availability of software for automated analysis, many products lack third party extensibility, the capability to perform simultaneous event detection and have no provision for anomaly detection in highly dense crowded scenes. We present a plugin based software system for video surveillance applications addressing these shortcomings and achieve realtime performance in typical crowded scenes. Core parameters are computed once per frame and shared among plugins to improve performance by eliminating redundant calculations. A novel multiple pedestrian tracking algorithm is incorporated into the framework to achieve the expected performance. We also propose an improvement to anomaly detection in densely crowded scenes using non-trajectory based dominant motion pattern clusters that can enhance the detection capability of the state-of-the-art.
item: Conference-Full-text
Faster human activity recognition with SVM
(2012) Chathuramali, KGM; Rodrigo, BKRP
Human activity recognition finds many applications in areas such as surveillance, and sports. Such a system classifies a spatio-temporal feature descriptor of a human figure in a video, based on training examples. However many classifiers face the constraints of the long training time, and the large size of the feature vector. Our method, due to the use of an Support Vector Machine (SVM) classifier, on an existing spatio-temporal feature descriptor resolves these problems in human activity recognition. Comparison of our system with existing classifiers using two standard datasets shows that our system is much superior in terms of the computational time, and either it surpasses or is on par with the existing recognition rates. It performs on par or marginally inferior to existing systems, when the number of training examples are a few due to the imbalance, although consistently better in terms of computation time.
item: Conference-Abstract
Feature enhancement for mean-Shift based object tracking
(2016-08-29) Gamage, DS; Samarakoon, B; Dabarera, R; Handagala, SM; Rodrigo, BKRP
In object tracking identifying the best feature which discriminates object and background improves the performance. Most of the existing methods do not consider the suitability of such features for the tracker. Here we enhance the discriminative features which elevate the tracker performance. To accommodate object and background variations over time we dynamically update the best feature using a distance measure. We demonstrate the performance of the resulting systems on the UNIVERSITATKARLSRUHE Image Sequences.
item: Conference-Abstract
FPGA-Based compact and flexible architecture for real-time embedded vision systems
Samarawickrama, M; Pasqual, AA; Rodrigo, BKRP
A single-chip FPGA implementation of a vision core is an efficient way to design fast and compact embedded vision systems from the PCB design level. The scope of the research is to design a novel FPGA-based parallel architecture for embedded vision entirely with on-chip FPGA resources. We designed it by utilizing block-RAMs and IO interfaces on the FPGA. As a result, the system is compact, fast and flexible. We evaluated this architecture for several mid-level neighborhood algorithms using Xilinx Virtex-2 Pro (XC2VP30) FPGA. Our algorithm uses a vision core with a 100 MHz system clock which supports image processing on a low-resolution image of 128×128 pixels up to 200 images per second. The results are accurate. We have compared our results with existing FPGA implementations. The performance of the algorithms could be substantially improved by applying sufficient parallelism.
item: SRC-Report
Multiple motion model computation in the presence of outliers
Rodrigo, BKRP
item: SRC-Report
Multiple Motion Model Computation in the Presence of Outliers
(2016-08-18) Rodrigo, BKRP
Real-time image processing demands much more processing power than a conventional processor can deliver. As a result hardware acceleration became necessary to augments processors with application-specific coprocessors. Due to the limited resources on 1;'l'GA and nature of some sequential algorithms, it is difficult to depend entirely on slice resources, In this research, we implemented a multiprocessor architecture to support reel-time image processing on FPGA.Furthermol'e, we benchmarked and compared our implemented architectures with their counterparts. The operational structure of multiprocessor architecture consists of on-chip processors implemented in a parallel manner with efficient memory and bus architectures, The performance properties such as accuracy. throughput and efficiency are measured and presented. Multiprocessor systems are effective in software level parallelism on FPGA. Our quad-Microblaze architecture achieved 75-80% performance improvement compared to its single Microblaze counterpart. Moreover, the quad-Microblaze design is faster than the single-powerPC implementation on l"PFA. Therefore, multi-processor architecture with customised coprocessors are effective for implementing custom parallel architecture to achieve real-time image processing,
item: Conference-Abstract
Patient alert and decision support system
Gunawardane, TSFW; Koggalage, R; Rodrigo, BKRP; Rajapakse, S
Safety of critically ill patients in intensive care units is an important aspect of medical care. There are many factors contributing to shortcomings and errors in patient care in the intensive care setting, such as long working hours, high levels of stress, lack of enough people, may cause human errors and affecting the effectiveness of the decisions of the physician. Several attempts have been made to increase the effectiveness of such decisions by issuing early alerts on adverse patient conditions. However, such alerts are based on single parameter variations but not on the relationship between multiple parameter variations. Thus, inability to provide an effective communication model causes a considerable bottleneck in intensive care unit (ICU) operations. The proposed model is an integrated solution which identifies the adverse patient conditions on multiple parameter variations and then provides predictive treatment suggestions on those identified conditions. It follows an interactive communication cycle in order to properly notify the responsible physicians. Results show that the system is capable of early identification of adverse conditions and providing suitable treatment suggestions compared to physicians themselves make decisions on same patient conditions.
item: Conference-Abstract
Recognition of Badminton strokes using dense trajectories
Ramasinghe, S; Chathuramali, KGM; Rodrigo, BKRP
Automatic stroke recognition of badminton video footages plays an important role in the process of analyzing players and building up statistics. Yet recognizing activities from broadcast videos is a challenging task due to person dependant body postures and blurring of the fast moving body parts. We propose a robust and an accurate approach for badminton stroke recognition using dense trajectories and trajectory aligned HOG features which are calculated inside local bounding boxes around players. A four-class SVM classifier is then used to classify badminton strokes to be either smash, forehand, backhand or other. This approach is robust to noisy backgrounds and provides accurate results for low resolution broadcast videos. Our experiments also reveal that this approach needs relatively fewer training samples for accurate recognition of strokes compared to existing approaches.
item: Article-Abstract
Robust and Efficient Feature Tracking for Indoor Navigation
Rodrigo, BKRP; Zouqi, M; Chen, Z; Samarabandu, J
Robust feature tracking is a requirement for many computer vision tasks such as indoor robot navigation. However, indoor scenes are characterized by poorly localizable features. As a result, indoor feature tracking without artificial markers is challenging and remains an attractive problem. We propose to solve this problem by constraining the locations of a large number of nondistinctive features by several planar homographies which are strategically computed using distinctive features. We experimentally show the need for multiple homographies and propose an illumination-invariant local-optimization scheme for motion refinement. The use of a large number of nondistinctive features within the constraints imposed by planar homographies allows us to gain robustness. Also, the lesser computation cost in estimating these nondistinctive features helps to maintain the efficiency of the proposed method. Our local-optimization scheme produces subpixel accurate featuremotion. As a result, we are able to achieve robust and accurate feature tracking
item: Conference-Abstract
Scanned–array audio beamforming using 2nd− and 3rd–order 2D IIR beam filters on FPGA
(2016-08-29) Ganganath, N; Attanayake, G; Bandara, TY; Ilangakoon, P; Rodrigo, BKRP; Madanayake, A; Bruton, LT
Real-time scanned-array direct-form-I hardware implementations of two-dimensional (2D) infinite impulse response (IIR) frequency-planar beam plane-wave (PW) filters have potentially wide applications in the directional enhancement of spatio-temporal broadband PWs based on their directions of arrival (DOAs). The proposed prototypes consist of a microphone sensor array, low-noise-amplifiers (LNAs), multiplexers (MUXs), a programmable gain amplifier (PGA), an analog to digital converter (ADC), a digital to analog converter (DAC), and a field programmable gate array (FPGA) circuit based 2D IIR spatiotemporal beam filter implemented on a single Xilinx Virtex2P xc2vp30-7ff896 FPGA chip. Starting from published 1st-order designs, novel FPGA architectures for highly-selective 2nd- and 3rd-order beam PW filters are proposed, simulated, implemented on FPGA, and verified on-chip.
item: Conference-Abstract
Shape prior based object extraction using a log-polar domian representation
Senarathna, J; Rodrigo, BKRP
In this paper, we address the problem of extracting objects in an image that conform to prior object shape knowledge. Major challenges include the judgment of scale and rotation parameters as well as tolerating occlusions and noise. We propose the use of a novel log polar domain mapping of the Cartesian domain image to efficiently and effectively overcome these. This method greatly simplifies rotation, scaling and provides an opportunity to incorporate a decision threshold. Whilst, the initial theoretical framework is developed on binary images, placement of this into the gray scale domain is achieved by incorporating a pre-processing binary segmentation step. A post processing Cartesian domain energy optimization is done to counteract discrepancies caused at the initial stage. We demonstrate our results using several examples.
item: Conference-Abstract
Use of a visual word dictionary for topic discovery in images
(2016-08-29) Kandasamy, K; Rodrigo, BKRP
The bag of visual words model has seen immense success in addressing the problem of image classification. Algorithms using this model generate the codebook of visual words by vector quantizing the features (such as SIFT) of the images to be classified. However, a codebook so formed tends to get biased by the nature of the dataset. In this paper we propose an alternative method to create the codebook for the dataset of images to be classified. Instead of directly using the dataset itself we first create a visual word dictionary by studying the SIFT features of a universal set of images. The codebook for the images to be classified is derived from this dictionary. To assess the effectiveness of the codebook thus derived, we classify the images using Probabilistic Latent Semantic Analysis in an unsupervised setting and Naive Bayes’ classification in a supervised setting. The use of a dictionary achieves results comparable to those obtained via a codebook formed from the dataset itself in much less computational time. We also use the dictionary to demonstrate how analogies can be drawn between visual words and linguistic words and present an analysis on one such analogy—that of polysemy.
item: Conference-Abstract
Vision based elephant recognition for management and conservation
(2016-08-29) Dabarera, R; Rodrigo, BKRP
In elephant management and conservation, it is vital to have non-invasive methods to track elephants. Image based recognition is a non-invasive mechanism for elephant tracking, albeit the inefficiency in manual method due to difficulties in handling large amount of data from multiple sources. To mitigate the drawbacks in manual method, we have proposed a computer vision based, automated, elephant recognition mechanism, which mainly relies on appearance based recognition algorithms. We have tested feasibility of the system running on a web based interface, which can facilitate researchers and conservationists all around the world to actively participate in elephant conservation.