Robust Fall Detection Using Human Shape and Multi-class Support Vector Machine

Foroughi, Homa; Rezvanian, Alireza; Paziraee, A.

doi:10.1109/icvgip.2008.49

Cited by 84 publications

(43 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, for the methods mentioned above, they either construct different models for different activities [4] and [6], or build a very complex structure to distinguish falls from other activities such as the multi-class SVM method in [5]. Our work is underpinned by the observation that the fall activity shares similarities and can be ascribed to one class.…”

Section: Introductionmentioning

confidence: 99%

“…C. Juang and C. Chang in [4] use an elegant self-constructing neural fuzzy inference network for posture recognition to detect a fall. As an effective tool for the classification problem, the SVM technique is applied in [5], the extracted features are finally fed to a multi-class SVM for precise classification of motions and determination of a fall event. In [6], a layered hidden Markov model (LHMM)-based approach is proposed to determine the state of the person (walking or falling) from a multiview pose classification strategy.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fall detection in a smart room by using a fuzzy one class support vector machine and imperfect training data

Naqvi

Rhuma

et al. 2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we propose an efficient and robust fall detection system by using a fuzzy one class support vector machine based on video information. Two cameras are used to capture the video frames from which the features are extracted. A fuzzy one class support vector machine (FOCSVM) is used to distinguish falling from other activities, such as walking, sitting, standing, bending or lying. Compared with the traditional one class support vector machine, the FOCSVM can obtain a more accurate and tight decision boundary under a training dataset with outliers. From real video sequences, the success of the method is confirmed with less non-fall samples being misclassified as falls by the classifier under an imperfect training dataset.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fall detection in a smart room by using a fuzzy one class support vector machine and imperfect training data

Naqvi

Rhuma

et al. 2011

2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Biometrics are automated methods of recognizing an individual based on their physiological (fingerprints, face, ear, iris) [1,2,3] or behavioral characteristics (gait, signature) [4]. Each biometric has its strengths and weaknesses and the choice typically depends on the application [5].…”

Section: Introductionmentioning

confidence: 99%

An Efficient Preprocessing Technique for Noise Reduction in Ear Verification System

Tavassoli¹,

Yaqubi²

2011

IJCA

View full text Add to dashboard Cite

Today, Biometric systems are considered superior in technological developments, because they provide a nontransferable means of identifying people not just cards or badges. The image enhancement step is designed to reduce noise in this area. The key point about an identification method that is "nontransferable" means it cannot be given or lent to another individual so nobody can get around the system they personally have to go through the control point. The image enhancement before feature extraction system can be very efficient. In this paper a new method is proposed to raise the performance of an ear verification system, since at first, using hybrid denoising method, the noises removed from ear image and then the next step denoisy image is used for verification system. Experimental results in this study show that Gaussian noises well removed from the ear images and has acceptable affect on verification accuracy.

show abstract

“…Instead of using 2-d features (as in [5], [6], [8] and [7]), we use 3-d features derived from various view angles of multiple cameras in order to reliably detect falls in different directions.…”

mentioning

confidence: 99%

“…In order to avoid the need to construct a model for each type of activity (such as walking, sitting, standing and crouching), as in [7], [8] and [9], we exploit a one-class classifier which captures the fall activity in a single class different from the non-fall classes. Our approach thereby has the advantage that training time and testing time for the proposed fall detection system will be reduced.…”

mentioning

confidence: 99%

One class boundary method classifiers for application in a video-based fall detection system

Naqvi

Rhuma

et al. 2012

IET Comput. Vis.

View full text Add to dashboard Cite

In this paper, we introduce a video-based robust fall detection system for monitoring an elderly person in a smart room environment. Video features, namely the centroid and orientation of a voxel person, are extracted. The boundary method, which is an example one class classification technique, is then used to determine whether the incoming features lie in the 'fall region' of the feature space, and thereby effectively distinguishing a fall from other activities, such as walking, sitting, standing, crouching or lying. Four different types of boundary methods, k-center, k-th nearest neighbor, one class support vector machine and single class minimax probability machine are assessed on representative test datasets. The comparison is made on the following three aspects: 1). True positive rate, false positive rate and geometric means in detection 2). Robustness to noise in the training dataset 3). The computational time for the test phase. From the comparison results, we show that the single class minimax probability machine achieves the best overall performance. By applying one class classification techniques with 3-d features, we can obtain a more efficient fall detection system with acceptable performance, as shown in the experimental part; besides, it can avoid the drawbacks of other traditional fall detection methods.

show abstract

Robust Fall Detection Using Human Shape and Multi-class Support Vector Machine

Cited by 84 publications

References 8 publications

Fall detection in a smart room by using a fuzzy one class support vector machine and imperfect training data

Fall detection in a smart room by using a fuzzy one class support vector machine and imperfect training data

An Efficient Preprocessing Technique for Noise Reduction in Ear Verification System

One class boundary method classifiers for application in a video-based fall detection system

Contact Info

Product

Resources

About