Bogdan Kwolek

Human fall detection on embedded platform using depth maps and wireless accelerometer

Since falls are a major public health problem in an aging society, there is considerable demand for low-cost fall detection systems. One of the main reasons for non-acceptance of the currently available solutions by seniors is that the fall detectors using only inertial sensors generate too much false alarms. This means that some daily activities are erroneously signaled as fall, which in turn leads to frustration of the users. In this paper we present how to design and implement a low-cost system for reliable fall detection with very low false alarm ratio. The detection of the fall is done on the basis of accelerometric data and depth maps. A tri-axial accelerometer is used to indicate the potential fall as well as to indicate whether the person is in motion. If the measured acceleration is higher than an assumed threshold value, the algorithm extracts the person, calculates the features and then executes the SVM-based classifier to authenticate the fall alarm. It is a 365/7/24 embedded system permitting unobtrusive fall detection as well as preserving privacy of the user.

Kernel particle filter for real-time 3D body tracking in monocular color images

This paper presents the application of a kernel particle filter for 3D body tracking in a video stream acquired from a single uncalibrated camera. Using intensity-based and color-based cues as well as an articulated 3D body model with shape represented by cylinders, a real-time body tracking in monocular cluttered image sequences has been realized. The algorithm runs at 7.5 Hz on a laptop computer and tracks the upper body of a human with two arms. First, experimental results show that the proposed approach has good tracking as well as recovering capabilities despite using a small number of particles. The approach is intended for use on a mobile robot to improve human robot interaction

Improving fall detection by the use of depth sensor and accelerometer

Since falls are a major cause of harm to older people, there is considerable demand for low-cost fall detection systems. To meet demands of the end-users we propose a new architecture for low cost and reliable fall detection, where an accelerometer is used to indicate a potential fall and the Kinect sensor is used to authenticate the eventual fall alert. In consequence, the depth maps are not processed frame-by-frame, but instead we download from a circular buffer a sequence of depth maps acquired prior to the fall and then process them to authenticate fall event. We determine features both in the depth maps and point clouds to extract discriminative fall descriptors. Since people typically follow typical motion patterns related to specific locations in home or typical daily activities, we propose to utilize k-nn classifier to implement an exemplar-based fall detector. We show that such a classifier is competitive on our publicly available URFD dataset in terms of sensitivity and specificity while being much more simple to implement on an embedded platform.

Fuzzy inference-based reliable fall detection using kinect and accelerometer

Falls are major causes of mortality and morbidity in the elderly. However, prevalent methods only utilize accelerometers or both accelerometers and gyroscopes to separate falls from activities of daily living. This makes it not easy to distinguish real falls from fall-like activities. The existing CCD-camera based solutions require time for installation, camera calibration and are not generally cheap. In this paper we show how to achieve reliable fall detection. The detection is done by a fuzzy inference system using low-cost Kinect and a device consisting of an accelerometer and a gyroscope. The experimental results indicate high accuracy of the detection and effectiveness of the system.

Face detection using convolutional neural networks and gabor filters

This paper proposes a method for detecting facial regions by combining a Gabor filter and a convolutional neural network. The first stage uses the Gabor filter which extracts intrinsic facial features. As a result of this transformation we obtain four subimages. The second stage of the method concerns the application of the convolutional neural network to these four images. The approach presented in this paper yields better classification performance in comparison to the results obtained by the convolutional neural network alone.

Multi camera person tracking applying a graph-cuts based foreground segmentation in a homography framework

Reliable tracking of objects is an inevitable prerequisite for automated video surveillance systems. As most object detection methods, which are based on machine learning, require adequate data for the application scenario, foreground segmentation is a popular method to find possible regions of interest. These usually require a specific learning phase and adaptation over time. In this work we will present a novel approach based on graph cuts, which outperforms most standard algorithms. It is commonly agreed that occlusions can only be resolved in multi camera environments. Applying multi layer homography will enable us to robustly detect and track objects applying only foreground data, resulting in a high tracking performance.

Fall detection on embedded platform using kinect and wireless accelerometer

In this paper we demonstrate how to accomplish reliable fall detection on a low-cost embedded platform. The detection is achieved by a fuzzy inference system using Kinect and a wearable motion-sensing device that consists of accelerometer and gyroscope. The foreground objects are detected using depth images obtained by Kinect, which is able to extract such images in a room that is dark to our eyes. The system has been implemented on the PandaBoard ES and runs in real-time. It permits unobtrusive fall detection as well as preserves privacy of the user. The experimental results indicate high effectiveness of fall detection.

Stereovision-Based Head Tracking Using Color and Ellipse Fitting in a Particle Filter

This paper proposes the use of a particle filter combined with color, depth information, gradient and shape features as an efficient and effective way of dealing with tracking of a head on the basis of image stream coming from a mobile stereovision camera. The head is modeled in the 2D image domain by an ellipse. A weighting function is used to include spatial information in color histogram representing the interior of the ellipse. The lengths of the ellipse’s minor axis are determined on the basis of depth information. The dissimilarity between the current model of the tracked object and target candidates is indicated by a metric based on Bhattacharyya coefficient. Variations of the color representation as a consequence of ellipse’s size change are handled by taking advantage of the scale invariance of the similarity measure. The color histogram and parameters of the ellipse are dynamically updated over time to discriminate in the next iteration between the candidate and actual head representation. This makes possible to track not only a face profile which has been shot during initialization of the tracker but in addition different profiles of the face as well as the head can be tracked. Experimental results which were obtained on long image sequences in a typical office environment show the feasibility of our approach to perform tracking of a head undergoing complex changes of shape and appearance against a varying background. The resulting system runs in real-time on a standard laptop computer installed on a real mobile agent.

Fuzzy inference-based fall detection using kinect and body-worn accelerometer

Graphical abstractDisplay Omitted HighlightsA new approach for reliable fall detection.In case of potential fall a threshold-based algorithm launches the fuzzy system to authenticate the fall event. The fuzzy system consists of two input Mamdani engines and a triggering alert Sugeno engine.The output of the first engine is a fuzzy set, which assigns grades of membership to the possible values of dynamic transitions, whereas the output of the second one is another fuzzy set assigning membership grades to possible body poses.Since the Mamdani engines perform fuzzy reasoning on disjoint subsets of the linguistic variables, the total number of the fuzzy rules needed for input-output mapping is far smaller. In this paper, we present a new approach for reliable fall detection. The fuzzy system consists of two input Mamdani engines and a triggering alert Sugeno engine. The output of the first Mamdani engine is a fuzzy set, which assigns grades of membership to the possible values of dynamic transitions, whereas the output of the second one is another fuzzy set assigning membership grades to possible body poses. Since Mamdani engines perform fuzzy reasoning on disjoint subsets of the linguistic variables, the total number of the fuzzy rules needed for input-output mapping is far smaller. The person pose is determined on the basis of depth maps, whereas the pose transitions are inferred using both depth maps and the accelerations acquired by a body worn inertial sensor. In case of potential fall a threshold-based algorithm launches the fuzzy system to authenticate the fall event. Using the accelerometric data we determine the moment of the impact, which in turn helps us to calculate the pose transitions. To the best of our knowledge, this is a new application of fuzzy logic in a novel approach to modeling and reliable low cost detecting of falls.

Articulated body motion tracking by combined particle swarm optimization and particle filtering

This paper proposes the use of a particle filter with embedded particle swarm optimization as an efficient and effective way of dealing with 3d model-based human body tracking. A particle swarm optimization algorithm is utilized in the particle filter to shift the particles toward more promising configurations of the human model. The algorithm is shown to be able of tracking full articulated body motion efficiently. It outperforms the annealed particle filter, kernel particle filter as well as a tracker based on particle swarm optimization. Experiments on real video sequences as well as a qualitative analysis demonstrate the strength of the approach.