المستودع الرقمي لـقسم هندسة البرمجيات

Antipersonnel landmines represent a very serious hazard endangering the lives of many people living in armed conflict counties. The huge number of human lives lost due to this phenomenon has been a strong motivation for this research. Deep Learning (DL) is considered a very useful tool in object detection, image classification, face recognition and other computer vision activities. This paper focuses on DL for the problem of landmines recognition in order to identify its type based on shape features. This research work consists of several stages: gathering a new dataset of Anti-Personnel Mines (APMs) images for training and testing purposes, employing several augmentation strategies to boost the diversity of training data, applying four different Convolutional Neural Network (CNN) models namely VGG, ResNet, MiniGoogleNet and MobileNet, and evaluating their performances on APMs recognition. In conclusion, results indicate that MiniGoogleNet exceed all of other three models in recognizing APMs with the highest accuracy rate of 97%.
Hassan Ali Hassan Ebrahem, Abdelhamid Elwaer, Marwa Solla, Fatima Ben Lashihar, Hala Shaari, Rudwan A. Husain(7-2021)

3D integral imaging is a true 3D imaging technology. It offers the simplest form that is capable of recording and replaying the true light field 3D scene in the form of a planar intensity distribution, by employing microlens array. There is a new world health crisis threatening the humanity with the spread of Covid-19 (Coronavirus Disease-2019). The Covid-19 belongs to a family of viruses that may cause various symptoms such as pneumonia, fever, breathing difficulty, and lung infection. Real images of Covid-19 patients confirmed by computed tomography CT were used to segment areas of increased attenuation in the lungs, all compatible with ground glass opacities and consolidations. This paper describes a new method to generate an indicative Covid-19 3D integral image models. The method is based on a Multiprocessor Ray Tracing System including Philips viewer, 3D slicer software and unidirectional camera. Experimental results are extremely satisfactory and for the first time it is proved that 3D integral images Covid-19 models are generated through Multiprocessor ray tracing system in order to deep monitoring and visualization to could be aid diagnosis in the absence of RT-PCR kits as demonstrated. A new file format content is created as well.
Dr. Mahmoud Geat Eljadid(3-2022)

3D Holoscopic imaging (also referred to as Integral Imaging) is a technique for creating full colour 3D optical models that exist in space independently of the viewer. In this paper, the main objective is to provide new file formats as raw material for full parallax computer generated 3D Holoscopic content, based mainly on Adapted Multiprocessor Ray Tracing System, displayed on auto-stereoscopic. The most important is being 3D model file format as there are many 3D modelling application available in the market and support different file formats. Therefore a file formats (scene description file) are developed for an Adapted Multiprocessor Ray Tracing System "Tachyon" (AMRTST) and for Ac3D computer graphics software package as well; the new file formats are created by uses 3D unidirectional camera parameters. Experimental results show validation of the new algorithm (AMRTST) and for the first time a 3D content of file formats are developed, and tests on an adapted Multiprocessor ray tracing system and on Ac3D package some scenes are tested such as teapot, car, roses and human-body.
Dr. Mahmoud Geat Eljadid, Prof. Amar Aggoun, Dr. Osama Hassan Youssef(3-2018)

3D Holoscopic imaging (also known as Integral imaging) is a technique that is capable of creating and encoding a true volume spatial optical model of the object scene in the form of a planar intensity distribution by using unique optical component This paper presents an new technique that can produce 3D integral imaging animations movies, using the integral ray tracing algorithm. The main purpose of this paper is to propose a plug-in tool to interface with commercially available Ac3D Computer Graphics software package. The proposed approach is based on a 3D integral imaging ray-tracer software, multiprocessor ray tracing system, and a 3D unidirectional integral camera model. Basically, 3D camera file is required to manipulate 3D camera viewing parameters in order to have a new position of the 3D integral imaging camera. Ac3D IIP is especially developed in order to read the Ac3D model file format. Experimental results show validation of the new adapted software tool and tests on such as car and roses models. Consequently, 3D holoscopic computer animation has been created. Although a basic 3D model was used, this still showcases the principles behind 3D holoscopic imaging and it has been successfully displayed on a HoloVizio display.
Dr. Mahmoud Geat Eljadid, Prof. Amar Aggoun(3-2018)

Holoscopic 3D imaging (H3D) technology also known as integral imaging is a true 3D imaging technology. It offers the simplest form that is capable of recording and replaying the true light field 3D scene in the form of a planar intensity distribution, by employing Microlens array. Despite it uses the same characteristics of holographic, it records the 3D information in 2D form and display in full 3D with optical component, without the need of coherent light source and confine dark fine. In addition, it facilitates postproduction processing such as refocusing. This makes it more practical approach for real-time 3D image capture and display. In this paper, a new method of extract 3D integral images data from raster maps. The R2V is an advanced raster to vector conversion software system is mainly used in order to generate geographical 3D integral images. Experimental results are extremely satisfactory and for the first time it is proved that, Geographic 3D Integral Images is generated. It means that, a new certain approach enables GIS applications of using 3D integral images is now established.
Dr. Mahmoud Geat Eljadid(12-2021)

Integral imaging, also known as Holoscopic imaging to be promising approach for glassless 3D and Its methodology uses the principle of “Fly’s eye” and hence allows natural viewing of objects (i.e. fatigue free viewing), In this paper, the main objective is to provide a new plug-in tools for full parallax computer generated 3D Holoscopic content, based mainly on 3D Virtual Reality Modeling Language parser to enable 3D integral images content and rendering to be produced from VRML file format, and displayed on auto-stereoscopic. In the proposed system, the 3D content is either captured by a single camera with a micro lens, cylindrical lens array or computer generated to allow for mixed 3D video generation. Multiprocessor ray tracing system is adapted to be able to generate 3D integral images that containing 3D IIVRML integral images content parser modules; new 3D IIVRML file format is created by uses 3D unidirectional camera parameters in order to import/accept into ray tracer renderer software. Experimental results show validation of the new plug-in software tool and tests on such Tie scene and Cessna scene . Consequently, 3D integral images frames and short time 3D II movie are generated and displayed without a glass on PC screen, LCD and the HoloVizio.
Dr. Mahmoud Geat Eljadid, Prof. Amar Aggoun(1-2018)

3D Holoscopic imaging (also known as Integral imaging) is a technique that is capable of creating and encoding a true volume spatial optical model of the object scene in the form of a planar intensity distribution by using unique optical components. It is akin to holography in that 3D information recorded on a 2D medium can be replayed as a full 3D optical model, Typically, moving from cylindrical lens to the next cylindrical lens is carried out knowing the lens width, Lw, which defines the translation of the camera along the x optical axis and hence the position of the new camera cylindrical lens. This new renderer is represented in the paper by the term integral ray tracing (IRT). In this paper, a method of generation 3D integral images camera file parameters is proposed. The camera file is an important part of the process of generating micro-images and set up the camera position. The method is mainly based on Parallel Multiprocessor Ray Tracing System adapted "Tachyon" as renderer. Experimental results show validation of the new algorithm and the lenticular lens can be set up and tested on small-balls, gears, room, tree and teapot scenes.
Dr. Mahmoud Geat Eljadid, Prof. Amar Aggoun, Dr. Osama Hassan Youssef(1-2018)

The main purpose of this paper is to introduce some techniques to overcome 3D missing information visual artefacts problem, holes and blank regions between 3D integral images (holoscopic images) frames encountered with the computer generated 3D integral image, pixel-tracing styles is proposed to achieve a higher performance of generating 3D integral images in terms of execution time, Z-buffer and correct occlusion scheme is also introduced to overcome visual artefact. The methods are based on a 3D integral images ray-tracer containing 3D integral images parser, 3D integral camera model, a 3D integral images renderer and scene transformations. Consequently, fast sequences of 3D integral image frames for animations purposes are generated, experimental results show validation of the strategies to fill in the 3D missing pixels and test on different scenes, improvement in terms of execution time depends on the complexity of the scene is achieved.
Dr. Mahmoud Geat Eljadid, Prof. Amar Aggoun, Dr. Osama Hassan Youssef(1-2014)