Web service implementation in the object detection on rural roads

Object detection is a challenge in the computer vision area. Traditional methods build from the extraction of their characteristics, to complex deep neural networks that combine object detection and classification. Currently, there are web services that allow the creation of models or methods that support the training and validation of the classifier to be built. In this article, we present the use of the web service Custom Vision by Azure, which is based on convolutional neural network architectures and helps developers in the construction of models that meet their requirements in the area of computer vision. The images to be evaluated are of roads in the rural area, where there is a geater complexity to detect objects. The Custom Vision service can be used to carry out classification or image detection tasks with some precision these will be presented in the results of the investigation.