Early Warning Device for Brown Plant Hopper Detection in Palay Using Wireless Sensor Networks

Rice (oryza sativa) is one of the most in demand staple foods in the world because of its versatility in processing and consumption. But due to the ever increasing human population, rice farmers and producers are pressured to produce more from just one to three cropping seasons. Climatic change also affects rice production as well as the heavy use of chemical inputs. These factors contribute to the problems of pest outbreaks and disease attacks to rice. Brown planthoppers (BPH, Nilaparvata lugens) are among the outbreak pests of rice that have brought heavy losses in several Asian countries including the Philippines. Its early detection is crucial because this pest is a virus carrier that can cause tremendous yield loss to rice. With this, designing an early warning device is a critical element for BPH pest management system. Image processing is one of the early detection mechanisms proposed in the literature. In this research, an early warning device using Wireless Sensor Network (WSN) through Haar feature-based cascade classifiers for BPH detection is being proposed which can be installed in strategic locations in the palay-farm-area under study. This device is programmed to send warning messages when acceptable range of the BPH cascade classifiers is obtained; thereby, allowing the end-users to check the status of the palay farm area. The researcher develops an early warning device prototype and performs an extensive set of experiments measuring its image detection capability.