Deep convolutional neural network (DCNN) based computer vision methods have great progress in object detection tasks. And object detection is essential to many applications such as autonomous robots and vehicles. Howe...
详细信息
ISBN:
(纸本)9781538675199
Deep convolutional neural network (DCNN) based computer vision methods have great progress in object detection tasks. And object detection is essential to many applications such as autonomous robots and vehicles. However, it is difficult for object detection system to be deployed on embedded platforms due to its intensive computing architecture. We present an object detector MobileNet-SSD which can be deployed on embedded platforms. The object detector presented in this work based on Single Shot Detector (SSD) framework and replace the feature extractor with a more light weight network MobileNetV1. General matrix to matrix multiplication is used to simplify the calculation of convolution operation and is further optimized using ARM NEON technology. Finally, the object detector MobileNet-SSD is deployed on embedded platform and achieves 1.13FPS with 0.72 mean average precision on PASCAL VOC dataset.
One of AI branch, Computer Vision-based recognition systems is necessary for security in Autonomous Vehicles (AVs). Traffic sign recognition systems are popularly used in AVs because it ensures driver safety and decre...
详细信息
ISBN:
(数字)9781728166094
ISBN:
(纸本)9781728166100
One of AI branch, Computer Vision-based recognition systems is necessary for security in Autonomous Vehicles (AVs). Traffic sign recognition systems are popularly used in AVs because it ensures driver safety and decrease vehicles accidents on roads. However, the inability of AVs to accurately detect road signs and pedestrian behaviour has led to road crashes and even death in recent times. Additionally, as cities become smarter, the traditional traffic signs dataset will change considerably, as theGoogle, 2020se vehicles and city infrastructure introduce modern facilities into their operation. In this paper, we introduce a computer vision based road surface marking recognition system to serve as an added layer of data source from which AVs will make decisions. We trained our detector using YOLOv3 running in the cloud to detect 25 classes of Road surface markings using over 25,000 images. The results of our experiment demonstrate a robust performance in terms of the accuracy and speed of detection. The results of which will consolidate the traffic sign recognition system, thereby ensuring more reliability and safety in AVs decision making. New algorithm using Deep Learning technology in Artificial intelligence (AI) application is implemented and tested successfully.
暂无评论