Image compression plays an important role in reducing the size and storage of data while increasing the speed of its transmission through the Internet significantly. Image compression is an important research topic for several decades and recently, with the great successes achieved by deep learning in many areas of image processing, especially image compression, and its use is increasing Gradually in the field of image compression. The deep learning neural network has also achieved great success in the field of processing and compressing various images of different sizes. In this paper, we present a structure for image compression based on the use of a Convolutional AutoEncoder (CAE) for deep learning, inspired by the diversity of human eyes' observation of the different colors and features of images. We propose a multi-layer hybrid system for deep learning using the unsupervised CAE architecture and using the color clustering of the K-mean algorithm to compress images and determine their size and color intensity. The system is implemented using Kodak and Challenge on Learned Image Compression (CLIC) dataset for deep learning. Experimental results show that our proposed method is superior to the traditional compression methods of the autoencoder, and the proposed work has better performance in terms of performance speed and quality measures Peak Signal To Noise Ratio (PSNR) and Structural Similarity Index (SSIM) where the results achieved better performance and high efficiency With high compression bit rates and low Mean Squared Error (MSE) rate the results recorded the highest compression ratios that ranged between (0.7117 to 0.8707) for the Kodak dataset and (0.7191 to 0.9930) for CLIC dataset. The system achieved high accuracy and quality in comparison to the error coefficient, which was recorded (0.0126 to reach 0.0003) below, and this system is onsidered the most quality and accurate compared to the methods of deep learning compared to the deep learning methods of the autoencoder
In this paper, an algorithm through which we can embed more data than the
regular methods under spatial domain is introduced. We compressed the secret data
using Huffman coding and then this compressed data is embedded using laplacian
sharpening method.
We used Laplace filters to determine the effective hiding places, then based on
threshold value we found the places with the highest values acquired from these filters
for embedding the watermark. In this work our aim is increasing the capacity of
information which is to be embedded by using Huffman code and at the same time
increasing the security of the algorithm by hiding data in the places that have highest
values of edges and less noticeable.
The perform
(3)
In this research a proposed technique is used to enhance the frame difference technique performance for extracting moving objects in video file. One of the most effective factors in performance dropping is noise existence, which may cause incorrect moving objects identification. Therefore it was necessary to find a way to diminish this noise effect. Traditional Average and Median spatial filters can be used to handle such situations. But here in this work the focus is on utilizing spectral domain through using Fourier and Wavelet transformations in order to decrease this noise effect. Experiments and statistical features (Entropy, Standard deviation) proved that these transformations can stand to overcome such problems in an elegant way.
... Show More
(4)
Background/Objectives: The purpose of this study was to classify Alzheimer’s disease (AD) patients from Normal Control (NC) patients using Magnetic Resonance Imaging (MRI). Methods/Statistical analysis: The performance evolution is carried out for 346 MR images from Alzheimer's Neuroimaging Initiative (ADNI) dataset. The classifier Deep Belief Network (DBN) is used for the function of classification. The network is trained using a sample training set, and the weights produced are then used to check the system's recognition capability. Findings: As a result, this paper presented a novel method of automated classification system for AD determination. The suggested method offers good performance of the experiments carried out show that the
... Show MoreAutomated detection of Dubas palm infestation by image processing techniques has practical significance as it can improve agricultural efficiency, increase crop yield and quality, protect the environment, and provide data-driven insights. It also reduces the human effort required for pest control and enhances sustainability. In this study, we aimed to automate the detection of Dubas bug infestation in palm trees using deep learning with transfer learning residual neural networks. Based on four models: InceptionResNetV2, ResNet18, ResNet50, and ResNet101, the data used in this study were obtained by drone photography, many images were taken, and then the infected area was extracted. Using two types of data, 185 infected images and 185 health
... Show More
(1)
(1)
Text based-image clustering (TBIC) is an insufficient approach for clustering related web images. It is a challenging task to abstract the visual features of images with the support of textual information in a database. In content-based image clustering (CBIC), image data are clustered on the foundation of specific features like texture, colors, boundaries, shapes. In this paper, an effective CBIC) technique is presented, which uses texture and statistical features of the images. The statistical features or moments of colors (mean, skewness, standard deviation, kurtosis, and variance) are extracted from the images. These features are collected in a one dimension array, and then genetic algorithm (GA) is applied for image clustering.
... Show More
(9)
(5)
(74)
(70)
Diabetic retinopathy is an eye disease in diabetic patients due to damage to the small blood vessels in the retina due to high and low blood sugar levels. Accurate detection and classification of Diabetic Retinopathy is an important task in computer-aided diagnosis, especially when planning for diabetic retinopathy surgery. Therefore, this study aims to design an automated model based on deep learning, which helps ophthalmologists detect and classify diabetic retinopathy severity through fundus images. In this work, a deep convolutional neural network (CNN) with transfer learning and fine tunes has been proposed by using pre-trained networks known as Residual Network-50 (ResNet-50). The overall framework of the proposed
... Show More
(5)
(5)