{"id":927,"date":"2026-07-08T19:43:18","date_gmt":"2026-07-08T19:43:18","guid":{"rendered":"https:\/\/abdullahjirjees.com\/projects\/"},"modified":"2026-07-21T19:59:24","modified_gmt":"2026-07-21T19:59:24","slug":"projects","status":"publish","type":"page","link":"https:\/\/abdullahjirjees.com\/ar\/projects\/","title":{"rendered":"Projects"},"content":{"rendered":"
\n\n

Applied AI & Computer Vision Projects<\/h1>\n\n\n

Selected research and engineering work in autonomous systems, robotics perception, and industrial computer vision.<\/p>\n\n\n

<\/div>\n\n<\/div>\n\n\n\n
\n\n
\n\n
\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

1. Multimodal Driver Fatigue & Cognitive Load Detection<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> Driver fatigue and cognitive overload are major safety risks in autonomous and semi-autonomous driving. Relying on a single sensor isn’t enough to catch it reliably.<\/p>\n\n\n

Approach:<\/strong> Built a multimodal AI pipeline combining visual signals (eye and mouth activity, pupil features) with physiological and vehicle-context data, using temporal deep learning (CNN-LSTM) to detect fatigue patterns over time rather than frame-by-frame.<\/p>\n\n\n

Outcome:<\/strong> A research-grade driver monitoring pipeline used in real-world experimental sessions. Contributed to applied autonomous-vehicle safety research and publication outputs.<\/p>\n\n\n

Python<\/code> PyTorch\/TensorFlow<\/code> OpenCV<\/code> CNN-LSTM<\/code> RealSense<\/code> CAN bus<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n\n

\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

2. ROS2-Based Multimodal Perception & Sensor Acquisition Platform<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> Autonomous-systems research needs reliable, synchronized data from many different sensors, without losing timing consistency or data quality.<\/p>\n\n\n

Approach:<\/strong> Designed a ROS2-based data acquisition workflow integrating cameras, stereo vision, thermal imaging, LiDAR, GNSS\/IMU, CAN bus, and physiological sensors for repeatable real-world data collection.<\/p>\n\n\n

Outcome:<\/strong> A structured multimodal data pipeline that improved repeatability and sensor organization, supporting autonomous systems experiments.<\/p>\n\n\n

ROS2<\/code> Python<\/code> OpenCV<\/code> ZED stereo<\/code> LiDAR<\/code> NVIDIA Jetson<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n<\/div>\n\n\n

\n\n
\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

3. UAV-Based Computer Vision for Object Detection & Visual Analytics<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> UAV imagery is hard to analyze automatically. Altitude, angle, lighting, and clutter all change constantly, and labeled data is limited.<\/p>\n\n\n

Approach:<\/strong> Built computer vision pipelines for detection, segmentation, and analytics from aerial imagery, including preprocessing, model training\/evaluation, and real-world field validation.<\/p>\n\n\n

Outcome:<\/strong> UAV-based AI workflows for real-world monitoring and inspection tasks, moving from research prototype toward operational use.<\/p>\n\n\n

Python<\/code> PyTorch<\/code> TensorFlow<\/code> OpenCV<\/code> NVIDIA Jetson<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n\n

\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

4. Real-Time Visual Scene Analysis for Autonomous Systems<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> Autonomous systems need to understand dynamic scenes, objects, lanes, road context, in real time and under real-world conditions.<\/p>\n\n\n

Approach:<\/strong> Developed a visual scene analysis pipeline combining object detection, tracking, and segmentation for real-time processing on autonomous vehicle research platforms.<\/p>\n\n\n

Outcome:<\/strong> A modular scene-understanding and tracking pipeline used to evaluate AI behavior under motion, viewpoint changes, and sensor noise.<\/p>\n\n\n

Python<\/code> OpenCV<\/code> YOLO-style detection<\/code> ROS2<\/code> ZED stereo<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n<\/div>\n\n\n

\n\n
\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

5. Industrial Computer Vision for High-Precision Visual Inspection<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> Industrial and semiconductor imaging requires reliably detecting fine visual defects and quality issues from high-resolution images.<\/p>\n\n\n

Approach:<\/strong> Developed image-processing and AI pipelines covering acquisition, preprocessing, feature extraction, and defect\/anomaly analysis, built for production reliability.<\/p>\n\n\n

Outcome:<\/strong> Contributed to production-grade computer vision systems for industrial inspection and metrology.<\/p>\n\n\n

Python<\/code> C\/C++<\/code> OpenCV<\/code> Defect detection<\/code> Git<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n\n

\n\n
\n\n
\n\n
<\/svg><\/div>\n\n\n

6. UAV & LiDAR-Based Pothole Detection and Depth Mapping<\/h3>\n\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Problem:<\/strong> Roads develop potholes that are costly and time-consuming to survey manually, and identifying both location and true depth is difficult at scale.<\/p>\n\n\n

Approach:<\/strong> Used a drone with a professional camera and a 64-channel Velodyne LiDAR to collect imagery and laser scan data, then trained a machine learning model for pothole detection and depth estimation on a GPU server.<\/p>\n\n\n

Outcome:<\/strong> The model successfully detected potholes and measured depth with strong accuracy, a scalable alternative to manual road-condition surveys.<\/p>\n\n\n

Python<\/code> PyTorch\/TensorFlow<\/code> LiDAR<\/code> UAV imagery<\/code> GPU training<\/code><\/p>\n\n<\/div>\n\n<\/div>\n\n<\/div>\n\n\n

Have a similar problem to solve? Book a Session<\/a> or Get in Touch<\/a>.<\/p>\n\n<\/div>","protected":false},"excerpt":{"rendered":"

Applied AI & Computer Vision Projects Selected research and engineering work in autonomous systems, robotics perception, and industrial computer vision. 1. Multimodal Driver Fatigue & Cognitive Load Detection Problem: Driver fatigue and cognitive overload are major safety risks in autonomous and semi-autonomous driving. Relying on a single sensor isn’t enough to catch it reliably. Approach: […]<\/p>\n","protected":false},"author":167,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_gspb_post_css":"","footnotes":""},"class_list":["post-927","page","type-page","status-publish","hentry"],"blocksy_meta":[],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t