Ermin Kevric

My research enhances fruit detection in orchards using advanced object detection models like Detectron2, YOLO, and vision transformers. I optimize tracking algorithms such as BoTSORT for apple tracking and leverage ROS and C++ for refining deep learning algorithms through multi-sensor fusion. Specifically, I have worked on LiDAR-camera sensor fusion for background filtering.

I researched the development and the application of deep learning algorithms towards anomaly detection and root cause analysis in time series.

I researched the application, and the improvement of neural network based computer vision algorithms, for automatic detection of CFRP defects. In addition, I learned about the development of an OPC UA server, using PROSYS Java SDK.

Through I-CCEW, I worked on commercializing a haptic feedback device for treating freezing of gait in Parkinson’s Disease. Particularly, I worked on interfacing Arduino with an IMU, programming it to detect 'frozen feet'. Our team followed an agile approach to create a mock-up in under 50 days, conducting over three dozen interviews. I have additionally researched the target market, helped create a business model, and attended 8 seminars on leadership, public speaking, and human-centered design"

I mentored 9 freshmen engineering students, drawing on my personal experience at the university, I provided guidance for academic and career success at the University of Oklahoma. As a mentor, I conducted 7 engagement activities, facilitated class adaptation, introduced career opportunities like the Career Fair, and shared insights from workshops. I also helped assign workstreams, oversaw two engineering design projects, and supported improving their presentation and communication skills

In addition to Major in Electrical Engineering, I have obtained a Minor degree in Mathematics