摘要
在本研究中發展雷射病蟲害防治儀，其結合了單眼立體視覺（monocular stereo vision）、快速雷射掃描（rapid laser scanning）與智慧影像辨識（intelligent pest recognition）等關鍵技術，應用於農業害蟲種群數量控制。選擇小白紋毒蛾（Orgyia Postica）與台灣黃毒蛾（Porthesia Taiwana）這兩種常見的害蟲做測試，將其標定完成之原始圖像用於YOLO演算法訓練與辨識，在YOLO演算法框選區域，再針對不同的害蟲做影像處理來檢測出害蟲的頭部，若是小白紋毒蛾則先需要將影像之色彩空間由RGB轉換成HSV後再進行頭部檢測的影像處理，並通過使用k均值聚類演算法（k-means clustering algorithm）定位出害蟲頭部的中心座標值，使雷射掃描系統能精準地控制雷射光束照射在害蟲頭部。實驗結果指出在波長450 nm、功率1.735 W、光斑直徑2.5 mm的條件下，二齡以上的台灣黃毒蛾幼蟲在照射1.2 sec後會使其無法進一步攝入食物。因此，所發展之雷射病蟲害防治儀是能夠有效地控制害蟲種群的方法。
關鍵字
無資料並列摘要
In this study, we developed a laser pest control system which combine the key techniques of monocular stereo vision, rapid laser scanning, and intelligent pest recognition for disabling caterpillars. Two caterpillar species of Orgyia Postica and Porthesia Taiwana were considered and their original images were used to train YOLO for identification. Color transform from RGB to HSV was applied to Orgyia Postica, and for Porthesia Taiwana, RGB color space was maintained, to successfully detect the caterpillar's head. The center of the caterpillar's head was positively approximated by using k-means clustering algorithm. These newly identified coordinates were then exposed to automatically controlled laser beam. Preliminary experimental results indicated that the second-instar larvae of Porthesia Taiwana could not further ingest food after 1.2 sec of irradiation under laser parameters with wavelength of 450 nm, power of 1.735 W and spot diameter of 2.5 mm. Therefore, this synergistic laser system seems to be a beneficial and promising approach to effectively control the pest population.