Person Identification Based on Micro-Doppler Signatures of Sit-to-Stand and Stand-to-Sit Movements Using a Convolutional Neural Network
20 Feb 2020-Vol. 4, Iss: 3, pp 1-4
TL;DR: The obtained results will prove that both the horizontal and vertical directions of the velocities of both movements include information that can be used to identify individuals, and this information can be obtained with micro-Doppler radar systems.
Abstract: This letter presents a method for person identification based on sit-to-stand and stand-to-sit movements using micro-Doppler radar measurements and a convolutional neural network (CNN). Two 24-GHz micro-Doppler radar systems placed directly above or behind participants will be used to measure the sit-to-stand and stand-to-sit movements of 10 participants. Images of the micro-Doppler signatures will be generated by subjecting the signals received by the radar to short-time Fourier transform. The generated images will then be used as input for the CNNs for training and evaluation purposes. The experiments verified the ability of the method to accurately identify people by measuring both their sit-to-stand and stand-to-sit movements. The identification accuracies for the sit-to-stand and stand-to-sit measurements were 93.6% and 94.9%, respectively, using the data of the radar placed above the participant, whereas the accuracy when placing the radar behind the participant was 92.9% for the sit-to-stand and 93.9% for the stand-to-sit movements. The obtained results will prove that both the horizontal and vertical directions of the velocities of both movements include information that can be used to identify individuals, and this information can be obtained with micro-Doppler radar systems.
Citations
More filters
TL;DR: In this article, the application of micro-Doppler radar (MDR) to gait classification based on fall risk-related differences using deep learning and gait parameter-based approaches was described.
Abstract: This paper describes the application of micro-Doppler radar (MDR) to gait classification based on fall risk-related differences using deep learning and gait parameter-based approaches. Two classification problems were considered in this study: elderly non-fallers and multiple fallers were classified to investigate the detection of fall risk-related gait differences, and middle-aged (50s) and elderly (70s) adults were classified to detect aging-related gait differences. The MDR signal data of the participants were simulated using an open motion capture gait dataset. The classification results obtained using the deep learning and gait parameter-based approaches showed that the classification accuracy achieved using a support vector machine with the gait parameters extracted from the MDR signals was better than that resulting from the deep learning of spectrogram (time-velocity distribution) images of the MDR signals for both classification problems. The gait parameter-based approach achieved the classification rates of 79 % for faller/non-faller classification and 82 % for 50s/70s classification, whereas the corresponding accuracies were 73 % and 76 %, respectively, using the deep learning approach. These results reveal that the gait parameters extracted via MDR measurements include sufficient information on gait to detect individuals with a high risk of falls and the gait parameter-based approaches are thus effective for both classification problems.
10 citations
TL;DR: The results suggest that combining sit-to-stand and stand- to-sit movements provides sufficient information for accurate person identification and such information can be remotely acquired using Doppler radar measurements.
Abstract: This article demonstrates the identification of 10 persons with 99% accuracy achieved by combining micro-Doppler signatures of sit-to-stand and stand-to-sit movements. Data from these movements are measured using two radars installed above and behind the person. Images of Doppler spectrograms generated using the measured data are combined and input to a convolutional neural network. Experimental results show the significantly better accuracy of the proposed method compared with conventional methods that do not perform data combination. The accuracy of identifying 10 participants having similar ages and physical features was 96–99%, despite the relatively small training set (number of training samples: 50–90 Doppler radar images per person). These results suggest that combining sit-to-stand and stand-to-sit movements provides sufficient information for accurate person identification and such information can be remotely acquired using Doppler radar measurements.
7 citations
Cites background or methods or result from "Person Identification Based on Micr..."
...This study proposes an accurate person identification method that combines the data from sit-to-stand and stand-tosit movements to improve the method introduced in [24]....
[...]
...Moreover, a comprehensive study on person identification using the micro-Doppler signatures of such movements and the performance improvement based on feature combinations of these movements have not been conducted including in our previous study [24]....
[...]
...Further, conduction of a more in-depth analysis than [24] reveals the detailed performance and effectiveness of person identification via sit-to-stand/stand-to-sit movements....
[...]
...The CNN for person identification has the ResNet-18 architecture [27], [28], because it was the most effective network for radar-based person identification in our previous study [24] and other recognition methods based on CNNs [17], [22]....
[...]
...The Doppler radar system to measure the sit-to-stand and stand-to-sit movements is similar to the one we used for [24]....
[...]
07 Jul 2011
TL;DR: This paper proposes a people identification method which uses weak evidences from pressure sensors, accelerometer sensors, and light sensors placed on a chair to recognize who is sitting on the chair without any psychological and physical burden on users.
Abstract: Abstract. This paper proposes a people identification method based on the sitting patterns. This method uses weak evidences from pressure sensors, accelerometer sensors, and light sensors placed on a chair to recognize who is sitting on the chair without any psychological and physical burden on users. We discuss how we have implemented the system using softmax regression model, gradient descent algorithm and nearest neighbor search algorithm. Our experimental result shows that this method can be used in places which has private properties such as a home or small a office.
TL;DR: In this paper , a smart wheelchair for fall risk assessment was developed, which automatically measures the time a user spends in sitting down and standing up, and a SVM (Support Vector Machine)-based regression filter was adopted.
References
More filters
27 Jun 2016
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.
123,388 citations
Proceedings Article•
03 Dec 2012TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Abstract: We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax. To make training faster, we used non-saturating neurons and a very efficient GPU implementation of the convolution operation. To reduce overriding in the fully-connected layers we employed a recently-developed regularization method called "dropout" that proved to be very effective. We also entered a variant of this model in the ILSVRC-2012 competition and achieved a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry.
73,978 citations
Proceedings Article•
01 Jan 2015TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.
49,914 citations
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.
42,067 citations
07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.
40,257 citations