The design of an FPGA based portable seizure detection system for epilepsy patients is presented... more The design of an FPGA based portable seizure detection system for epilepsy patients is presented in this paper. For complete hardware support, an FPGA based epilepsy detection system exploiting Artificial Neural Network (ANN) based classifier with on chip training and testing support have been presented in this paper. With the help of such Neural Network processor, the prevailing problems in software like latency can be overcome and, also one can proceed towards an ASIC design of such system which is in high demand in recent days in the field of neuromorphic computing. The proposed hardware design takes the epileptic seizure characteristics namely—mean, variance, skewness, and, kurtosis as the input vectors to the ANN model and generates the classification output similar to traditional software system but now with complete and isolated hardware support. The total framework which incorporates both training and testing has been designed using Verilog HDL on Xilinx Vivado software. Th...
A healthy blood pressure (BP) is essential for survival, and a continuous change can lead to seri... more A healthy blood pressure (BP) is essential for survival, and a continuous change can lead to serious health problems. Fatal diseases such as cerebral necrosis, kidney failure, cardiovascular disorders, hypertension, and so on, are often linked to abnormal blood pressure. Hypertension is one of the leading causes of death in the globe. For an early diagnosis and prevention of deadly occurrences, an effective technique for continuous BP monitoring is required. Therefore, for earlier clinical detection, a convolutional neural network (CNN) based model for regular BP monitoring have been presented in this paper. The data used for this purpose is taken from the University of Guilin's photoplethysmograph (PPG).
Traditional von Neumann architecture based processors become inefficient in terms of energy and t... more Traditional von Neumann architecture based processors become inefficient in terms of energy and throughput as they involve separate processing and memory units, also known as~\textit{memory wall}. The memory wall problem is further exacerbated when massive parallelism and frequent data movement are required between processing and memory units for real-time implementation of artificial neural network (ANN) that enables many intelligent applications. One of the most promising approach to address the memory wall problem is to carry out computations inside the memory core itself that enhances the memory bandwidth and energy efficiency for extensive computations. This paper presents an in-memory computing architecture for ANN enabling artificial intelligence (AI) and machine learning (ML) applications. The proposed architecture utilizes deep in-memory architecture based on standard six transistor (6T) static random access memory (SRAM) core for the implementation of a multi-layered perce...
3D color channel based adaptive contrast enhancement using compensated histogram system
Multimedia Systems, 2021
Low contrast image is one of the major challenges in photography. The low contrast image not only... more Low contrast image is one of the major challenges in photography. The low contrast image not only poses difficulty to the interpretation of the scene but also causes trouble in the onward processing of the image for computer vision tasks. Histogram equalization (HE) is a traditional and widely used approach for contrast enhancement and applies to almost all types of images. However, HE causes over-enhancement of the image which degrades its natural appearance. In this paper, a novel scheme for enhancing the image contrast while retaining its naturalness has been proposed. The proposed method uses the compensated histogram equalization technique on each channel individually followed by blending of the channels with a suitable adaptive brightness adjustment kernel. The three color channels are combined to form the intermediate image. The high-frequency noise introduced during the process is filtered out. Finally, adaptive power law transformation is applied to adjust the overall brightness and to retain its naturalness. This makes the method strong in terms of contrast enhancement along with details preservation. The proposed method is applicable to all the contrast degraded images as it automatically adjusts its parameters based on the degradation level. The simulation results, on the CSIQ dataset, show that the proposed method performs better, qualitatively, and quantitatively than the existing methods.
Uploads
Papers by Abhash Kumar