病蟲害乃是造成溫室作物經濟損害的重要因素之一。由於在溫室環境下，作物緊密種植在有限的區域之中，一旦其中有一作物遭受病蟲害感染，其餘作物皆難逃其害。因傳統監測並計算害蟲數量之方法過於耗費人力成本，於本篇論文中，我們提出了一套基於影像分割與分類的自動化溫室小型害蟲辨識技術，目的為辨識出黑翅蕈蠅、薊馬類、粉蝨類三類蘭園常見之溫室小型害蟲並將成果應用於蘭園害蟲時空模式分析之中。本研究之實驗方法為掃描置於蘭園溫室之黃色黏蟲紙，透過影像分割技術偵測黏蟲紙影像上的蟲隻邊界並取出蟲隻體型大小及色彩資訊特徵，而後使用這些特徵進行三種目標害蟲之分類，其成果顯示該技術之演算法於600 dpi解析度下擁有較好的成果 (F1-score: 0.77)，且每張影像辨識所需時間約為40秒，符合蘭園害蟲辨識需求。此外，本次試驗於南臺灣之蘭花溫室進行約五個月份之黏蟲紙資料收集，運用辨識演算法成果描繪每週之三類小型害蟲空間分佈並進行熱點分析，此成果可輔助非農業及昆蟲專業背景之蘭園業主自主分析蟲害密集區進行精準用藥，毋須仰賴專業背景知識與長期經驗。

The pests are important economic loss factors for greenhouse. The plants in greenhouse are usually cultivated densely in small areas, so the infected plants may affect other plants easily. The existing detection method requires a significant amount of human labors to count the numbers of insects. In this study, we proposed an image based automatic detection method targeting for three common small-sized pest species (Fungus gnats, thrips and whitefly) in orchid greenhouse and the spatial-temporal analysis were applied to the results. The proposed method finds the boundary of insects and extract the body size and color features from hue component, which is calculated from the scanning image of sticky traps; afterwards, these features can be used to classify detected insects into corresponding species. The proposed method performed better with the optical resolutions of 600 dpi (F1-score: 0.77), compared to 200, 400, and 1200 dpi. The computation time is around 40 second for each image with the optical resolution of 600 dpi. The weekly spatial-temporal maps and hot spot analysis were materialized for an orchid greenhouse, located in southern Taiwan, for a period of 5 months.

Birch, C. P., Oom, S. P., & Beecham, J. A. (2007). Rectangular and hexagonal grids used for observation, experiment and simulation in ecology. Ecological Modelling, 206(3), 347-359.
Chen, J., & Lin, M. (2006). Developing Automatically Orchid Plants Biological Seneing and Cutivated Techniques (in Chinese). AGRICULTURE Policy & Review, 168, 86-91.
Chung, B. K., Xia, C. L., Song, Y. H., Lee, J. M., Li, Y., Kim, H., & Chon, T. S. (2014). Sampling of Bemisia tabaci adults using a pre-programmed autonomous pest control robot. Journal of Asia-Pacific Entomology, 17(4), 737-743. doi:10.1016/j.aspen.2014.06.009
Dallot, S., Gottwald, T., Labonne, G., & Quiot, J.-B. (2003). Spatial pattern analysis of Sharka disease (Plum pox virus strain M) in peach orchards of southern France. Phytopathology, 93(12), 1543-1552.
Fawcett, T. (2006). An introduction to ROC analysis. Pattern recognition letters, 27(8), 861-874.
Gonzalez, R. C., & Woods, R. E. (2006). Digital Image Processing (3rd Edition): Pearson.
Larios, N., Deng, H. L., Zhang, W., Sarpola, M., Yuen, J., Paasch, R., . . . Dietterich, T. G. (2008). Automated insect identification through concatenated histograms of local appearance features: feature vector generation and region detection for deformable objects. Machine Vision and Applications, 19(2), 105-123. doi:10.1007/s00138-007-0086-y
Legendre, P., & Fortin, M. J. (1989). Spatial pattern and ecological analysis. Vegetatio, 80(2), 107-138.
Liu, T., Chen, W., Wu, W., Sun, C. M., Guo, W. S., & Zhu, X. K. (2016). Detection of aphids in wheat fields using a computer vision technique. Biosystems Engineering, 141, 82-93. doi:10.1016/j.biosystemseng.2015.11.005
Martineau, M., Conte, D., Raveaux, R., Arnault, I., Munier, D., & Venturini, G. (2017). A survey on image-based insect classification. Pattern Recognition, 65, 273-284. doi:10.1016/j.patcog.2016.12.020
Miranda, J. L., Gerardo, B. D., & Tanguilig III, B. T. (2014). Pest detection and extraction using image processing techniques. International Journal of Computer and Communication Engineering, 3(3), 189.
Mitchell, A. (2005). The ESRI Guide to GIS Analysis, Volume 2: Spatial Measurements and Statistics: ESRI Press.
Ord, J. K., & Getis, A. (1995). Local spatial autocorrelation statistics: distributional issues and an application. Geographical analysis, 27(4), 286-306.
Solis-Sanchez, L. O., Castaneda-Miranda, R., Garcia-Escalante, J. J., Torres-Pacheco, I., Guevara-Gonzalez, R. G., Castaneda-Miranda, C. L., & Alaniz-Lumbreras, P. D. (2011). Scale invariant feature approach for insect monitoring. Computers and Electronics in Agriculture, 75(1), 92-99. doi:10.1016/j.compag.2010.10.001
Solis-Sanchez, L. O., Garcia-Escalante, J. J., Castaneda-Miranda, R., Torres-Pacheco, I., & Guevara-Gonzalez, R. (2009). Machine vision algorithm for whiteflies (Bemisia tabaci Genn.) scouting under greenhouse environment. Journal of Applied Entomology, 133(7), 546-552. doi:10.1111/j.1439-0418.2009.01400.x
Wen, C. L., & Guyer, D. (2012). Image-based orchard insect automated identification and classification method. Computers and Electronics in Agriculture, 89, 110-115. doi:10.1016/j.compag.2012.08.008
Wen, C. L., Guyer, D. E., & Li, W. (2009). Local feature-based identification and classification for orchard insects. Biosystems Engineering, 104(3), 299-307. doi:10.1016/j.biosystemseng.2009.07.002
Wen, T.-H., Lin, M.-H., Teng, H.-J., & Chang, N.-T. (2015). Incorporating the human-Aedes mosquito interactions into measuring the spatial risk of urban dengue fever. Applied Geography, 62, 256-266.
Xia, C. L., Chon, T. S., Ren, Z. M., & Lee, J. M. (2015). Automatic identification and counting of small size pests in greenhouse conditions with low computational cost. Ecological Informatics, 29, 139-146. doi:10.1016/j.ecoinf.2014.09.006
Xia, C. L., Lee, J. M., Li, Y., Chung, B. K., & Chon, T. S. (2012). In situ detection of small-size insect pests sampled on traps using multifractal analysis. Optical Engineering, 51(2). doi:10.1117/1.oe.51.2.027001