Keep the balance between high productivity, produce quality food and yet manage efficiently the resources available for production is one of the biggest challenges in the agricultural sector. As a possible solution to overcome these challenges and obstacles, precision agriculture emerged. Among the promising solutions that precision agriculture offers highlights the use of edge computing devices for monitoring and acquiring data in the rural environment, processing information locally and in real time. Computer Vision and Artificial Intelligence, more specifically Deep Learning, have also being applied recently in agriculture for different tasks such as image classification and object detection and semantic segmentation. However, there is a challenge and limitation of transferring this technology to more affordable platforms to process the data. Therefore, in this work, it was explored the use of computer vision and Deep Learning applied to the object detection task in edge devices, specifically the Raspberry PI 4 platform, without hardware acceleration. It was decided to apply this methodology for weed detection, once weeds are currently one of the pests that most cause loss of productivity in agriculture, also develop resistance for most of the herbicides used commercially. Also, in order to evaluate the performance gain for real-time weed detection on the Raspberry Pi platform, quantization of the deep neural network architecture using TensorFlow Lite was tested. The experimental results point out that the proposed methodology is functional, being possible to reproduce this experiment on similar edge devices, real time object detection was also achieved for the Raspberry Pi 4 platform.