Ikram Hattab

Skills: Zabbix · Java · Spring MVC · SQL · jaxio · Primefaces · Maven · Git · Jetty · Jasper Reports · Jasper

- Developing a model to detect salient regions with R3Net and Visual Salient Transformer (VST). - Implementing an Intelligent Bot capable of serving as a conversational assistant for ordering pizza (RASA). - Realize application to monitor cable test table with PYQT. - Development of an E-commerce website with Angular/Node. - Development of an E-commerce website with WordPress.

- Diagnosis and Optimization of Websites and E-commerce - Know how we can optimize a website - SEO, SEA, SMO, and SEM - Create an e-commerce website using Shopify and WordPress

- Web application which allows you to manage compressed air leak detection campaigns, to map the sources of leaks, to quantify energy losses and centralize repair actions (Angular , Node and MongoDB) - Dedicated platform for municipal officers to track peoples' claims (Angular, Node and MongoDB) - Develop doctor platform to track patients health ( Angular, Node and MongoDB) - Develop an ECG signal processing system model for the detection of cardiac disorders(Python)

Robotics vision: Object detection from visible and thermal cameras In our end-of-study internship, we focus on the problem of automatic detection by robots for surveillance applications. Particularly, the main goal to perform detection in a bi-spectral vision. Since deep learning frameworks have recently achieved great success in many applications mainly in the field of computer vision, we intend in this project to make use of such networks in order to obtain good results for the detection task. Using inputs from visible and thermal cameras, we propose a fusion network that uses features from both of them. This fusion process is incorporated into YOLO-v3 architecture. The resulting custom model of the detector is trained in a paired setting in order to improve the results compared to the detection of each input independently. To show the effectiveness of the proposed fusion framework, some experiments are conducted on KAIST, a multi-spectral dataset well-known in the field and commonly used for detection. Also, the detection process is applied in real-time on PGuard robot manufactured by Enova ROBOTICS, the hosting company of this project. In total, this work can be summarized in two steps. The first step is to develop a detection model based on a feature fusion method, then the second step is to integrate the object detection application model in PGuard Robot. Keywords :Robotic Vision, Deep Learning, Neural Networks, Object Detection, YOLO,Visible and Thermal Cameras, Saliency Map, Fusion, bi-spectral vision

Real estate management website (Laravel ): allows the customer to rent a place that fits his needs by identifying details for each type of real estate. Used Laravel Developed both backend and frontend

Invoice management website (Laravel ): allows the company manager to create invoices for his customers and send them by email or print them out. Used Laravel Developed both backend and frontend

Implementation and development of plugins in WordPress (Php): The project consists in developing plugins in WordPress that allow to: manage user profiles, manage publications shared by users, ensure communication between users, messaging. Used PHP Developed both backend and frontend

Customer files management (JavaSwing): facilitates the communication between customers and employees: it allows to organize the tasks of employees requested by the customers (installations, realizations of Web site..). Used Java swing