초록

용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

Responding quickly to the occurrence of pests in farms is an important issue that minimizes the damage to the farmers and prevents the pest damage from spreading to other farms. In this paper, we propose an integrated system for mobile-based pest detection, which can search for pests in real-time a...

Responding quickly to the occurrence of pests in farms is an important issue that minimizes the damage to the farmers and prevents the pest damage from spreading to other farms. In this paper, we propose an integrated system for mobile-based pest detection, which can search for pests in real-time at farms and conduct self-diagnosis based on the field. With existing systems that diagnose or predict existing pests, user can make a quick diagnostic on site. Studies related to pest detection have been limited to only a limited number of pests, and showed low level of accuracy in pest recognition. The integrated system for pest detection proposed by this paper adopts both similarity based image retrieval and recognition technology using deep learning algorithm. Similarity-based image retrieval techniques have the advantage of being able to rapidly apply to various pests of crops, but they show relatively low accuracy. On the other hand, image recognition based image retrieval technology requires a lot of data to learn. So it takes considerable time and it is difficult to apply to various pests but it shows relatively high accuracy. Thus, using both techniques, both of the diversity and high accuracy of pest awareness were explored and complemented. In addition, this system can detect the pest in real time at the farm using the mobile device, making the self-diagnosis through the search.

주제어