This paper presents a new configurable pruning Gaussian image filter CMOS architecture to address energy efficiency requirements regarding edge detection applications. Low-energy consumption is key for Internet of Thi...
This paper presents a new configurable pruning Gaussian image filter CMOS architecture to address energy efficiency requirements regarding edge detection applications. Low-energy consumption is key for Internet of Things (IoT) devices. Many emerging IoT applications rely on cameras to extract video or image features by running power-hungry computer vision algorithms. The Gaussian image filter is one of the most compute intensive tasks for pre-processing edge detection techniques which are widely adopted in the computer vision domain. Therefore, our proposed 2D Gaussian filter architecture enables: i) a low power and low area overhead runtime configuration scheme based on clock gating technique to prune the Gaussian filter (GF) window size, and ii) run-time capability to balance the tradeoff between edge detection quality and energy efficiency. Our proposed configurable architecture is synthesized and mapped onto 45 nm technology for an ASIC implementation. Results show that for 6 different run-time profiles our proposed configurable architecture provides power dissipation reduction of up to 64% with multiple levels of edge detection quality, which is assessed by considering the performance conformance metric.
With the growing evolution of wireless communication technologies, there is still a need for higher data rates, increased system capacity, and improved service quality. OFDM WiMAX technology is now regarded as one of ...
With the growing evolution of wireless communication technologies, there is still a need for higher data rates, increased system capacity, and improved service quality. OFDM WiMAX technology is now regarded as one of the most common solutions for Broadband Wireless Connectivity in Urban Areas, capable of offering faster implementation and lower costs than standard wired options. This paper proposes effective adaptive algorithm processing with MMSE for use in wireless networks based on SISO and MIMO OFDM WiMAX, enabling network performance to be enhanced in the case of non-LOS wireless communications, which are standard in urban conditions. On the performance of the system, signal attenuation, the effects of several paths l, different mobility speeds and Doppler shift were studied. Combines the adaptive algorithm with MMSE, achieves improved joint channel estimation and signal detection which performs the technique effectively mobile. SNR, MSE and noise components are used to analyses mathematical models of adaptive modulation for transmitting images in SISO and MIMO systems. Simulation results show that the adaptive algorithm with MMSE would improve throughput. For example, when SNR equal 15 dB, the probability of MSE for BPSK based on MIMO principle is equal to 0.0016 with adaptive algorithm. Also, for the same value of SNR, the probability of MSE for BPSK based on MIMO principle is equal to 0.164 without adaptive algorithm. It can also be concluded that when processing signals in a receiving system under conditions of multi-path signal propagation, the use of adaptive algorithms with MMSE has a positive effect on noise immunity.
Human evaluation of SAR is time-consuming and costly. Typically it requires the indirect usability-based assessment of SAR system components or SAR systems from which the images arose. We investigate an assessment sys...
详细信息
ISBN:
(纸本)9782874870538
Human evaluation of SAR is time-consuming and costly. Typically it requires the indirect usability-based assessment of SAR system components or SAR systems from which the images arose. We investigate an assessment system which aims at finding digital signal processingalgorithms to simulate, complement and partly replace the human evaluation of SAR images. To better understand the human evaluation, expert SAR interpreters have been asked to solve tasks on SAR images whose different image qualities result from a specific SAR system by varying the parameter settings of one SAR system component. The SAR system component investigated first is the coding system where the spatial and the amplitude resolution are fundamental parameters. In this paper, we describe first results of a human evaluation with expert interpreters where the two coding standards JPEG and HEVC intra coding were evaluated at different spatial resolutions and data rates. The SAR image quality preferred to work with was identified by the interpreters.
Energy-efficient computer vision is vitally important for embedded and mobile platforms where a longer battery life can allow increased deployment in the field. In image sensors, one of the primary causes of energy ex...
详细信息
ISBN:
(数字)9781728143002
ISBN:
(纸本)9781728143019
Energy-efficient computer vision is vitally important for embedded and mobile platforms where a longer battery life can allow increased deployment in the field. In image sensors, one of the primary causes of energy expenditure is the sampling and digitization process. Smart subsampling of the image-array in a manner that is task-specific, can result in significant savings of energy. We present an adaptive algorithm for video subsampling, which is aimed at enabling accurate object detection, while saving sampling energy. The approach utilizes objectness measures, which we show can be accurately estimated even from sub-sampled frames, and then uses that information to determine the adaptive sampling for the subsequent frame. We show energy savings of 18 - 67% with only a slight degradation in object detection accuracy in experiments. These results motivated us to further explore energy-efficient subsampling using advanced techniques such as, reinforcement learning and Kalman filtering. The experiments using these techniques are underway and provide ample support for adaptive subsampling as a promising avenue for embedded computer vision in the future.
Breast cancer is one of the most typical sorts of cancer worldwide, and the most frequent cancer in women. Mammography remains the most effective tool for the early detection of breast cancer, as well as the systems o...
详细信息
ISBN:
(数字)9781728108469
ISBN:
(纸本)9781728108476
Breast cancer is one of the most typical sorts of cancer worldwide, and the most frequent cancer in women. Mammography remains the most effective tool for the early detection of breast cancer, as well as the systems of computer-aided diagnosis (CAD) is typically used as a second opinion by the radiologists. These CAD systems based on mammography images for automatic diagnosis breast cancer. Generally, the mammography images contain some noises and additional unnecessary objects, Such as: artifacts, identification labels, etc. These objects can disrupt the diagnostic process partially or completely. Therefore, a technique of preparing these images before using it is mandatory. So, in this article, we propose a method for automatic preprocessingimage. Our method consists of five steps. Firstly suppressed the additional unnecessary objects, next by removing the additional black background, third find the orientation of mammograms, fourth segmented the pectoral muscle, and we end by contrast enhancement. The proposed method is tested on DDSM (Digital database for Screening Mammography) database, consisting of 10480 mammograms, and Mini-MIAS (Mammogram image Analysis Society, UK) database, consisting of 322 mammograms. The obtained results confirm the efficacy of the proposed method.
Unconsciousness is a state which takes place when the person is unable to maintain his/her awareness of the surroundings. It is a fully or nearly loss of responsiveness to environmental stimuli. As a consequence of th...
详细信息
ISBN:
(纸本)9781450364287
Unconsciousness is a state which takes place when the person is unable to maintain his/her awareness of the surroundings. It is a fully or nearly loss of responsiveness to environmental stimuli. As a consequence of that state, physical injuries, and or mental problems might take place. Hence, this paper proposes a simple faint detection technique using Faster Region-based Convolutional Neural Network (Faster RCNN) with thermal imaging camera to observe the unconscious person or a fainted one, and consequently an instant and convenient treatment will be carried out to that person. Experimental results reveal that the proposed architecture achieved a precise accuracy during the dim lightening condition, and a fairly excellent result indoor in ascertaining where the case is.
This article presents an analysis of methods used for circle detection in an image along with a description of the newly proposed method of circle detection in a pre-processed image based on trigonometric functions so...
详细信息
This article presents an analysis of methods used for circle detection in an image along with a description of the newly proposed method of circle detection in a pre-processed image based on trigonometric functions so that the algorithm itself is feasible for low-power digital measuring systems. The method is based on the use of significant points found in the object that are connected with a line for which directions are calculated using the algorithm. From the information thus obtained, the geometric centre of the object can be determined using the algorithm. The developed method is compared with the classical method of detecting circles in an image using Hough transforms. Finally, the test results are stated in the article. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
The traditional human motion perception and behavior recognition methods are based on machine vision and imageprocessing. This method has the disadvantages of strong intrusion of individual data privacy, large enviro...
详细信息
The traditional human motion perception and behavior recognition methods are based on machine vision and imageprocessing. This method has the disadvantages of strong intrusion of individual data privacy, large environmental impact and tedious data processing, and cannot meet the practical requirements in real time. This paper introduced a solution of accurately capture human motion by using multi-sensor of MEMS to construct Zigbee wireless network, and utilizing SVM classification algorithm to realize human action recognition efficiently. The system is used to detect the abnormal human fall behavior, which is perfect reliability and real-time performance and the recognition accuracy of the action is above to 90%.
Cache replacement policies in chip multiprocessors (CMP) have been investigated extensively and proven able to enhance shared cache management. However, competition among multiple processors executing different thread...
详细信息
Cache replacement policies in chip multiprocessors (CMP) have been investigated extensively and proven able to enhance shared cache management. However, competition among multiple processors executing different threads that require simultaneous access to a shared memory may cause cache contention and memory coherence problems on the chip. These issues also exist due to some drawbacks of the commonly used Least Recently Used (LRU) policy employed in multiprocessor systems, which are because of the cache lines residing in the cache longer than required. In imageprocessing analysis of for example extra pulmonary tuberculosis (TB), an accurate diagnosis for tissue specimen is required. Therefore, a fast and reliable shared memory management system to execute algorithms for processing vast amount of specimen image is needed. In this paper, the effects of the cache replacement policy in a partitioned shared cache are investigated. The goal is to quantify whether better performance can be achieved by using less complex replacement strategies. This paper proposes a Middle Insertion 2 Positions Promotion (MI2PP) policy to eliminate cache misses that could adversely affect the access patterns and the throughput of the processors in the system. The policy employs a static predefined insertion point, near distance promotion, and the concept of ownership in the eviction policy to effectively improve cache thrashing and to avoid resource stealing among the processors.
Advanced Driver-Assistance systems (ADAS) can help drivers in the driving process and increase the driving safety by automatically detecting objects, doing basic classification, implementing safeguards, etc. ADAS inte...
详细信息
ISBN:
(纸本)9781538678855
Advanced Driver-Assistance systems (ADAS) can help drivers in the driving process and increase the driving safety by automatically detecting objects, doing basic classification, implementing safeguards, etc. ADAS integrate multiple subsystems including object detection, scene segmentation, lane detection, and so on. Most algorithms are now designed for one specific task, while such separate approaches will be inefficient in ADAS which consists of many modules. In this paper, we establish a multi-task learning framework for lane detection, semantic segmentation, 2D object detection, and orientation prediction on FPGA. The performance on FPGA is optimized by software and hardware co-design. The system deployed on Xilinx zu9 board achieves 55 FPS, which meets real-time processing requirement.
暂无评论