Embedded video processing on Raspberry Pi

Árva, Gábor; Fryza, Tomáš

doi:10.1109/radioelek.2017.7937598

Cited by 4 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The OpenCV software library uses image analysis to recognize the various types of information in an image, such as: detection of hand gestures, as set out in [23]; human facial recognition [24], where in this particular paper [25] the authors implemented a prototype using the Java CV library [26], which analyzes the camera stream from a IP security camera and detects human presence; and recognition and extraction of vehicle registration information [27] or surveillance security systems [28], in which the authors coupled common web cameras to devices of small processing power, e.g., Raspberry Pi, which acquires the image from the camera and uses a cloud platform to process the image for movement detection.…”

Section: Related Workmentioning

confidence: 99%

“…Case Study Why Was Chosen [23] Hand gesture detection Image data extraction using OpenCV [24] Video processing on Raspberry PI OpenCV on Raspberry PI [25] Human facial recognition OpenCV on Raspberry PI [27] Extraction of vehicle information OpenCV on Raspberry PI [28] Surveillance system Image acquisition on Raspberry PI [29] Indoor/outdoor person detection Uses RFID tags to detect humans [30] Indoor human detection Uses infrared sensor to detect humans [31] Indoor monitoring Uses IoT devices with AI IndoorCare Indoor human monitoring Uses IoT devices with Computer Vision…”

Section: Referencementioning

confidence: 99%

“…In Figure 6, on the right side, the threshold which corresponds to the movement detected on the left side of the image is displayed, with the movement area defined by a blue rectangle. To effectively calculate the threshold, as stated in [24], it is necessary at an early stage to convert the color image to a grayscale image, so the only differentiating factor is the pixel brightness. Then, it is necessary to blur the image so that there are no sudden changes in the pixel tones.…”

Section: System Operationmentioning

confidence: 99%

“…Figure 7 shows the different types of blurs supported by OpenCV: Gaussian Blur, Median Blur, and Normalized Block Blur [48]. To effectively calculate the threshold, as stated in [24], it is necessary at an early stage to convert the color image to a grayscale image, so the only differentiating factor is the pixel brightness. Then, it is necessary to blur the image so that there are no sudden changes in the pixel tones.…”

Section: System Operationmentioning

confidence: 99%

See 3 more Smart Citations

IndoorCare: Low-Cost Elderly Activity Monitoring System through Image Processing

Fuentes

Correia

Costa

et al. 2021

Sensors

View full text Add to dashboard Cite

The Portuguese population is aging at an increasing rate, which introduces new problems, particularly in rural areas, where the population is small and widely spread throughout the territory. These people, mostly elderly, have low income and are often isolated and socially excluded. This work researches and proposes an affordable Ambient Assisted Living (AAL)-based solution to monitor the activities of elderly individuals, inside their homes, in a pervasive and non-intrusive way, while preserving their privacy. The solution uses a set of low-cost IoT sensor devices, computer vision algorithms and reasoning rules, to acquire data and recognize the activities performed by a subject inside a home. A conceptual architecture and a functional prototype were developed, the prototype being successfully tested in an environment similar to a real case scenario. The system and the underlying concept can be used as a building block for remote and distributed elderly care services, in which the elderly live autonomously in their homes, but have the attention of a caregiver when needed.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Referencementioning

confidence: 99%

Section: System Operationmentioning

confidence: 99%

Section: System Operationmentioning

confidence: 99%

See 2 more Smart Citations

IndoorCare: Low-Cost Elderly Activity Monitoring System through Image Processing

Fuentes

Correia

Costa

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Another study by Jagannathan et al [14] ran a 7-layer CNN on TDA3x SoC for object classification, and the overall system performance was 15fps. The majority of existing work (Ratnayake et al [15], Bako et al [16], Nikitakis et al [17], Chen et al [18], Arva et al [19]) used only a single embedded device to conduct video processing, which is obviously not capable of fulfilling real-time video processing (24fps). Gao et al [20] built an embedded cluster for video processing, but it lacked a complete solution to address system stability and fault tolerance.…”

Section: Introductionmentioning

confidence: 99%

A Streaming Cloud Platform for Real-Time Video Processing on Embedded Devices

Zhang

Zhao

et al. 2021

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

Real-time intelligent video processing on embedded devices with low power consumption can be useful for applications like drone surveillance, smart cars, and more. However, the limited resources of embedded devices is a challenging issue for effective embedded computing. Most of the existing work on this topic focuses on single device based solutions, without the use of cloud computing mechanisms for parallel processing to boost performance. In this paper, we propose a cloud platform for real-time video processing based on embedded devices. Eight NVIDIA Jetson TX1 and three Jetson TX2 GPUs are used to construct a streaming embedded cloud platform (SECP), on which Apache Storm is deployed as the cloud computing environment for deep learning algorithms (Convolutional Neural Networks-CNNs) to process video streams. Additionally, self-managing services are designed to ensure that this platform can run smoothly and stably, in the form of a metric sensor, a bottleneck detector and a scheduler. This platform is evaluated in terms of processing speed, power consumption, and network throughput by running various deep learning algorithms for object detection. The results show the proposed platform can run deep learning algorithms on embedded devices while meeting the high scalability and fault tolerance required for real-time video processing.

show abstract