1,720,972 research outputs found
High Performance Embedded Short Time Fourier Transform Architecture for Real-Time Speech Enhancement using Differential Microphone Arrays
No full textDifferential Microphone Arrays (DMAs) are devices that exploit the differences in Time of Arrivals (ToA) of the input audio between each sensor in order to separate sources from different directions. Those devices have been used for many different purposes, like audio source localization, speech enhancement, noise suppression etc. DMA is very attractive for processing broadband signals such as speech because of their frequency-invariant beampatterns and small dimension, which make them suitable for the integration into wearable devices such as hearing aids, smartphones and headphones.
However, DMA white noise gain (WGN), especially at low frequencies, limits the array performance in terms of obtainable signal-to-noise gain. This limitation can be overcame by processing the audio signal in the Short-Time Fourier transform (STFT) domain at the price of higher complexity of the signal-processing unit. In order to achieve the desired high data throughput for real time applications, the STFT filterbank structure has been implemented in a Cortex M4F embedded CPU using SMID instruction and an embedded Floating Point Unit. The DMA beamforming coefficients can be calculated from the desired beampatterns once offline, stored in system memory and then applied to each STFT frame in order to reduce computational cost.
Combining hardware speedups, specialized SMID DSP instructions, high performance FPU and coefficients pre-calculation, the described architecture can achieve 4.096Mbps with only a 8ms of latency, making it suitable for real-time applications
Embedded Systems and TensorFlow Frameworks as Assistive Technology Solutions
No full textIn the field of deep learning, this paper presents the design of a wearable computer vision system for visually impaired users. The Assistive Technology solution exploits a powerful single board computer and smart glasses with a camera in order to allow its user to explore the objects within his surrounding environment, while it employs Google TensorFlow machine learning framework in order to real time classify the acquired stills. Therefore the proposed aid can increase the awareness of the explored environment and it interacts with its user by means of audio messages
Bioimpedance based monitoring system for people with neurogenic dysfunction of the urinary bladder
No full textPatients with impaired bladder volume sensation have the necessity to monitor bladder level in order to avoid urinary tract infections and urinary reflux that can lead to renal failure. In this paper the the effectiveness of an embedded and wearable solution for bladder volume monitoring using the bioimpedance measurement is tested. Data are streamed real-Time using Bluetooth wireless technology. The bioimpedance measurements on a healthy subject prove the effectiveness of the proposed solution. In the future the system will be evaluated in real world scenarios with patients affected by spinal paralysis and bladder neurogenic dysfunction
Using tensorflow to design assistive technologies for people with visual impairments
No full textTensorFlow is an open source deep learning framework developed at Google that enables developers to conceive a wide variety of applications based on artificial intelligence principles. In this paper, we employ such software resources towards the development of a computer vision system for people with visual impairments. We propose a wearable assistive technology solution consisting of a single board computer connected to a camera mounted on the user's glasses. A TensorFlow based software runs on the board in order to real time classify the images captured by the camera, while a text to speech process vocalizes the still's content for the blind person. In this way, the system provides an audio description of the objects in the user's surrounding environment and it may help these people to better detect the things around them
Scene Reconstruction from a Single Depth Image Using 3D CNN
No full textScene reconstruction from multiple viewpoints are not always possible and rather it represents a small minority of the potential applications, from robotic manipulators to drones, autonomous vehicles etc... To overcome those limitations, we propose a fully convolutional 3D neural network capable of reconstructing a full scene from a single depth image by creating a 3D representation of it and automatically filling holes and inserting hidden elements. Our algorithm was evaluated on a real word dataset of tabletop scenes acquired using a Kinect and processed using KinectFusion software in order to obtain ground truth for network training and evaluation. Extensive measurements show that our deep neural network architecture outperforms the previous state of the art in terms of both precision and recall for the scene reconstruction task
Area and power consumption trade-off for Σ-Δ decimation filter in mixed signal wearable IC
No full textArea and Power consumption are important design metrics in integrated circuit (IC), in particular in those targeted for wearable devices. Σ-Δ Analog to Digital Converter (ADC) are increasing in popularity in those devices thanks to the low bandwidth of a great number of sensors that permits to increase converter performances by the oversampling and noise shaping techniques. One of the most important part of the Σ-Δ ADC is the decimation filter, usually implemented as a Cascaded — Integrator — Comb (CIC). The various CIC architectures, in particular the Recursive and Non recursive — Polyphase ones, are well known in literature. However, filters on-chip performances are strictly related to the effective implementations. The aim of this paper is to evaluate the two architectures, with different values of the characteristic parameters, optimizing the −180 nm CMOS Standard Cell technology — design for a reduced area occupation or power consumption. Results prove that polyphase implementations, differently from theoretical analysis, are generally more power efficient than the recursive one only in a clock gated design, even with a higher area occupation. In addition, an estimation of the power consumption is provided using least squares regression
Embedded implementation of an eye-in-hand visual servoing control for a Wheelchair Mounted Robotic Arm
No full textWheelchair Mounted Robotic Arms (WMRA) can be used by people with severe motor skill impairment, such as SMA (Spinal Muscular Atrophy), Cerebral Palsy etc⋯, in order to achieve daily life tasks. Many of those systems have been presented in literature and are available on the market but they are really expensive and bulky. Instead we propose a simple robotic arm with 4 Degrees of Freedom (DOF), controlled by a cheap embedded platform in order to keep the size and the cost as low as possible. This paper presents the design and evaluation of a simple visual servoing control loop for the robotic arm based on a camera mounted on the arm end effector. A lot of attention has been given to the design process of an effective and more accessible human machine interface, in order to make the arm simply usable by the user. Tests and evaluations were performed on a simplified version of the final system prove the effectiveness of the proposed solution. However further optimization have to be achieved in order to make the whole system really usable in a real-word scenario
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Wearable speech enhancement system based on MEMS microphone array for disabled people
No full textDisabled people, especially the ones with motor skill impairments, have difficulties in interacting with personal computers and smartphones. Indeed Automatic Speech Recognition (ASR) could be helpful for those people, but it's limited in scenarios not affected by environmental noise that can decrease performance of the recognition, limiting user experience. We propose a speech enhancement system based on MEMS microphone array and a digital signal processor in order to increase signal-to-noise ratio (SNR) of the user's voice. The audio delay between microphones is exploited by the array using the Differential Microphone Array (DMA) and an Adaptive Noise Reduction techniques. In such way the system can obtain an increment in SNR about 16.5 dB, when noise and voice come from opposite directions. A voice activity detection (VAD) block recognizes when the user speaks and sends the data to a cloud-based ASR system. Due to the small array size, the embedded system can be integrated in a wearable device. Theoretical analysis and in-system measurements prove the effectiveness of the proposed solution
- …
