Offline Handwritten Text Recognition (HTR) is a technique for translating handwritten images into digitally editable text format. Due to the presence of cursive letters, punctuation marks, and compound characters, it is more complex to recognize Bangla handwritten text. Over the years, several approaches to the optical model of the HTR system have been developed, including Hidden Markov Model (HMM...
Histogram Equalization is one of the most common algorithms used in image processing applications. Traditional software based histogram equalization using CPU is no longer sufficient for high-computation based real-time image processing applications. To meet high computation demand, image processing operations can be hardware accelerated using Field Programmable Gate Array (FPGA). In this paper, we implemented histogram equalization using Verilog Hardware Description Language (HDL) on Xilinx FPGA for hardware acceleration. The result shows that our FPGA-based histogram equalization is on average 1.5 thousand times faster than traditional software based histogram equalization on CPU.
Video compression is frequently used for reducing the original data in order to store or communicate the data through a transmission line. All modern video codecs use the flexible block partitioning at the time of rate-distortion (RDO) process. As a result, these codecs minimize the bits but has a immense time complexity. So, computational time complexity is one the main problems for video encoders and decoders. In this work, we proposed a machine learning approach called gradient descent algorithm for fast CU splitting to reduce the time complexity of the video encoder by using the concept of skip criterion. Experimental results show that on average 44.18% encoding time can be saved over HEVC (High Efficiency Video Coding). On the other hand, on average, 1.34% bit rate is increased as a trade-off.
Coronavirus disease or COVID-19 is one of the most frightening and infectious diseases of the twenty-first century. Since the outbreak of COVID-19 in Wuhan, China, numerous researches are conducted in this sector. At the preliminary stage, there was not sufficient numeric data for research but when we consider the text data such as trending topics of Social Media or patients sharing experiences ab...
Autism is a set of complicated developmental disorders marked by social skills impairments, communication difficulties (verbal and nonverbal), and recurring behavior. Autistic children are frequently alienated as a result of these impairments. Rapid recognition of autism can help to establish a treatment strategy and lessen the burden on sufferers. As a result, effective methods to early diagnosis and treatment for ASD are necessary. The toddler, child, adolescent, and adult screening datasets are collected in this study and separated according to gender (male and female). By using random oversampling (ROS), these datasets are balanced. Next, different classifiers are applied to both the primary and balanced datasets. The MLP classifier produced the best results, and the hyperparameter for it was tuned to improve autism identification rate. However, the experimental outcome for the female dataset is better than the male dataset. The shapely adaptive explanation (SHAP) method is also employed to assess the significant features of male and female.
