Dr. Jwan Najeeb Saeed

Participate as an author ... Read More

Abstract: A fundamental goal of digital image compression is to reduce the bit rate for transmission or data storage while maintaining an acceptable fidelity or image quality. In this study a proposed coding (compression) scheme for grey scale image by combining discrete wavelet transform (DWT), multistage vector quantization (MSVQ) and Huffman coding is presented. These combinations are utilized to take the advantages provided by all of them to get high compression ratio with acceptable recovered image quality in term of Peak Signal to Noise Ratio (PSNR) instead of using each method separately. First, the discrete wavelet transform is performed on the original image using bi-orthogonal 9/7 filter (bior4.4) resulting in number of sub-bands according to the decomposition level that can be one, two or three in the proposed scheme. Huffman coding applied on the approximation sub-band, and on the indices of the last level detail sub-bands that are vector quantized by two stages using LBG algorithm which forms the basis of most vector quantizer designs, other detail sub-bands set to zero if the decomposition level is greater than one to increase the compression ratio. ... Read More

The conference was held virtually at Zoom plateform ... Read More

2022 Second International Conference on Computer Science, Engineering and Applications (ICCSEA) ... Read More

—While beauty is subjective, it is not easy to quantify. Assessing facial beauty based on a computer perspective is an emerging research area with various applications. Different trainable models have been proposed to identify the attractiveness of facial beauty utilizing different types of features, machine learning techniques and lately, convolutional neural networks (CNNs) have proven their efficiency in image classification. The main objective of recent previous work is to enhance the performance of the existing trainable methods and make them suitable for beauty attractiveness identification. In this study, the accuracy and effectiveness of four affective pre-trained CNNs models (AlexNet, GoogleNet, ResNet-50, and VGG16) in assessing the attractiveness of human facial images using the CelebA dataset have been explored, evaluated, and analyzed. The results demonstrate that GoogleNet surpassed the investigated pretrained networks with a performance accuracy of 82.8%. ... Read More