應用影像辨識技術判定水果成熟之研究

極低的農民平均所得與農村青壯年人口出走都市，促使農村勞動人口高齡化，再受全球人口受生育率下降之影響（少子化來臨），原農村生產人力在不足情況下，使的農業生產力大受衝擊。另外，我國地理環境欠佳與農事全面機械化成本過高的壓力，讓農業生產成本相對偏高，競爭性低。因此，唯有提升農業生產力、強化產業結構調整以及科技創新研發一途，才能根本解決源頭棘手問題。本研究於此思維下，設計開發一套水果成熟影像辨識系統，以提供農民與相關之研究人員進行深度使用。本研究設計一套應用先進的電腦視覺水果成熟辨識系統，藉由擷取影像鏡頭所拍攝畫面，辨識出畫面中的水果是否成熟。並將數據回傳給系統，以利進行後續處理，為生產力4.0科技農業開啟先端，為此研究之貢獻。

關鍵字

水果成熟；電腦視覺；辨識系統；生產力4.0

並列摘要

Young people who live in rural areas usually move to large cities to make money because of low average farm income. In the country, farmers are getting older. The cause of population aging is owing to the population birth rate descent, so the agriculture population becomes small. On the other hand, the agricultural production cost in Taiwan is relatively high while tackling the outside competition. To deal with this situation, we must upgrade our techniques, making the best use of IT to improve agriculture productivity. To help farmers and the researchers, we propose in this paper a computer vision system which can be used to check whether fruits are mature or not. Recognition system gives some feedback to the person responsible for further processing. This study also contributes to the agriculture technology as well as the Productivity 4.0 that is the pioneer of the former.

並列關鍵字

fruits mature ； computer vision ； recognition system ； productivity 4.0

參考文獻

行政院農業委員會(2018)。水果王國。存取日期：2017 年 1 月 20 日，取自https://www.coa.gov.tw/search_wg.php?search_index_group=1&search_index_class=&q=%E6%B0%B4%E6%9E%9C%E7%8E%8B%E5%9C%8B

Google Scholar

Wang, N., and Yeung, D. Y. (2013). Learning a deep compact image representation for visual tracking. In Proceeding of Advances in Neural Information Processing Systems 2013, 809-817.