Abnormal Crops Image Data Acquisition Strategy by Exploiting Edge Intelligence and Dynamic-Static Synergy in Smart Agriculture

Abstract

Abnormal crops image data play crucial role in controlling crop diseases and pest for smart agriculture. However, current agricultural image acquisition methods suffer from low-value data. This article presents a new strategy for collect high-quality image data for abnormal crops. First, a novel agricultural Internet of Things (IoT) image acquisition system is proposed, that integrates edge intelligence, motion–static synergy, which enables both coarse and fine crop image acquisition. To enhance image acquisition efficiency and data value in the agricultural IoT, this article proposes an image acquisition method based on edge intelligence and static and motion collaboration, using banana plantations as the example object. The method comprises three phases. In the first phase, the edge server deploys the YOLO-FDAC target detection model to detect abnormal crops from the images captured by static nodes. In the second phase, the coordinate solution method of abnormal crops and the quantification method of the degree of abnormality is presented. In the third phase, based on the severity of abnormality and the ant colony optimization, a path optimization algorithm for the image acquisition robot is designed. Finally, this article evaluates the performance of each level of the proposed method by comparing it with traditional methods. The experimental results demonstrate that the proposed image acquisition strategy has high acquisition efficiency and high image data value.

Keywords

• Edge intelligence
• image acquisition
• motion collaboration
• path optimization target detection