Performance investigation of two-stage detection techniques using traffic light detection dataset

<span>Using a camera to monitor an object or group of objects over time is the process detection. It can be used for variety things, including security and surveillance, video communication, traffic light detection (TLD), from compressed in public places. In recent times, tracking has become popular topic computer science particularly, data community, thanks usage deep learning (DL) artificial intelligence (AI). DL which convolutional neural network (CNN) as one its techniques usually two-stage methods TLD. Despite all successes recorded TLD through use methods, there no study that analyzed these experimental research, studying strength witnesses by researchers. Based on needs this analyses applications We implemented using 5 with dataset Jupyter notebook sklearn libraries. present achievements TLD, going standard performance metrics used, FASTER-CNN was best accuracy, F1-score, precision recall 0.89, 0.93, 0.83 0.90 respectively.</span>