General Background: Deep image matting is a fundamental task in computer vision, enabling precise foreground extraction from complex backgrounds, with applications in augmented reality, computer graphics, and video processing. Specific Background: Despite advancements in deep learning-based methods, preserving fine details such as hair and transparency remains a challenge. Knowledge Gap: Existing approaches struggle with accuracy and efficiency, necessitating novel techniques to enhance matting precision. Aims: This study integrates deep learning with fusion techniques to improve alpha matte estimation, proposing a lightweight U-Net model incorporating color-space fusion and preprocessing. Results: Experiments using the AdobeComposition-1k dataset demonstrate superior performance compared to traditional methods, achieving higher accuracy, faster processing speed, and improved boundary preservation. Novelty: The proposed model effectively combines deep learning with fusion techniques, enhancing matting quality while maintaining robustness across various environmental conditions. Implications: These findings highlight the potential of integrating fusion techniques with deep learning for image matting, offering valuable insights for future research in automated image processing applications, including augmented reality, gaming, and interactive video technologies. Highlights: Better Precision: Fusion techniques enhance fine detail preservation. Faster Processing: Lightweight U-Net improves speed and accuracy. Wide Applications: Useful for AR, gaming, and video processing. Keywords: Deep image matting, computer vision, deep learning, fusion techniques, U-Net
The field of autonomous robotic systems has advanced tremendously in the last few years, allowing them to perform complicated tasks in various contexts. One of the most important and useful applications of guide robots is the support of the blind. The successful implementation of this study requires a more accurate and powerful self-localization system for guide robots in indoor environments. This paper proposes a self-localization system for guide robots. To successfully implement this study, images were collected from the perspective of a robot inside a room, and a deep learning system such as a convolutional neural network (CNN) was used. An image-based self-localization guide robot image-classification system delivers a more accura
... Show MoreIts well known that understanding human facial expressions is a key component in understanding emotions and finds broad applications in the field of human-computer interaction (HCI), has been a long-standing issue. In this paper, we shed light on the utilisation of a deep convolutional neural network (DCNN) for facial emotion recognition from videos using the TensorFlow machine-learning library from Google. This work was applied to ten emotions from the Amsterdam Dynamic Facial Expression Set-Bath Intensity Variations (ADFES-BIV) dataset and tested using two datasets.
This paper presents a nonlinear finite element modeling and analysis of steel fiber reinforced concrete (SFRC) deep beams with and without openings in web subjected to two- point loading. In this study, the beams were modeled using ANSYS nonlinear finite element
software. The percentage of steel fiber was varied from 0 to 1.0%.The influence of fiber content in the concrete deep beams has been studied by measuring the deflection of the deep beams at mid- span and marking the cracking patterns, compute the failure loads for each deep beam, and also study the shearing and first principal stresses for the deep beams with and without openings and with different steel fiber ratios. The above study indicates that the location of openings an
This paper deals with proposing new lifting scheme (HYBRID Algorithm) that is capable of preventing images and documents which are fraud through decomposing there in to the real colors value arrays (red, blue and green) to create retrieval keys for its properties and store it in the database and then check the document originality by retrieve the query image or document through the decomposition described above and compare the predicted color values (retrieval keys) of the query document with those stored in the database. The proposed algorithm has been developed from the two known lifting schemes (Haar and D4) by merging them to find out HYBRID lifting scheme. The validity and accuracy of the proposed algorithm have been ev
... Show MoreAbstract The Object of the study aims to identify the effectiveness of using the 7E’s learning cycle to learn movement chains on uneven bars, for this purpose, we used the method SPSS. On a sample composed (20) students on collage of physical education at the university of Baghdad Chosen as two groups experimental and control group (10) student for each group, and for data collection, we used SPSS After collecting the results and having treated them statistically, we conclude the use 7E’s learning cycle has achieved remarkable positive progress, but it has diverged between to methods, On this basis, the study recommended the necessity of applying 7E’s learning cycle strategy in learning the movement chain on uneven bar
... Show MoreThe aim of this study to identify the effect of using two strategies for active learning ( Jigsaw Strategy & Problems Solving) in learning some balanced beam's skills in artistic gymnastics for women , as well as to identify the best of the three methods (jigsaw strategy , problems solving and the traditional method) in learning some skills balance beam , the research has used the experimental methodology, and the subject included the students of the college of Physical Education and Sports Sciences / University of Baghdad / third grade and by the lot was selected (10) students for each group of groups Search three and The statistical package for social sciences (SPSS) was used means, the standard deviation and the (T.test), the one way a n
... Show MoreThe literature shows conflicting outcomes, making it difficult to determine how e-learning affects the performance of students in higher education. The effect of e-learning was studied and data has been gathered with the utilization of a variety of qualitative and quantitative methods, especially in relation to students' academic achievements and perceptions in higher education, according to literature review that has been drawn from articles published in the past two decades (2000-2020). The development of a sense of community in the on-line environment has been identified to be one of the main difficulties in e-learning education across this whole review. In order to create an efficient online learning community, it could be claim
... Show MoreIn all applications and specially in real time applications, image processing and compression plays in modern life a very important part in both storage and transmission over internet for example, but finding orthogonal matrices as a filter or transform in different sizes is very complex and importance to using in different applications like image processing and communications systems, at present, new method to find orthogonal matrices as transform filter then used for Mixed Transforms Generated by using a technique so-called Tensor Product based for Data Processing, these techniques are developed and utilized. Our aims at this paper are to evaluate and analyze this new mixed technique in Image Compression using the Discrete Wavelet Transfo
... Show More