Local feature-based identification and classification for orchard insects

Pose estimation-dependent identification method for field moth images using deep learning architecture

“…Note that 6% of Argyrotaenia velutinana samples were misidentified to be Platynota idaeusalis samples because of the highly similarity of the shape and size between two species. In agreement with this result, Wen et al (2009) reported similar results of mis-identification between the Argyrotaenia b i o s y s t e m s e n g i n e e r i n g 1 3 6 ( 2 0 1 5 ) 1 1 7 e1 2 8 velutinana and Platynota idaeusalis samples. Eleven percent of Grapholita molesta samples were mis-identified to be Cydia pomonella samples because of the highly similarity of the texture feature between the two species.…”

“…Eleven percent of Grapholita molesta samples were mis-identified to be Cydia pomonella samples because of the highly similarity of the texture feature between the two species. Similarly, 3% of Platynota idaeusalis samples were mis-identified to be Argyrotaenia velutinan samples because of the highly similarity of the shape and size between the two species (Wen et al, 2009) and 3% of Platynota idaeusalis samples were mis-identified to be Spilonota ocellana samples because of the textural similarity of two species.…”

Geostatistical tools for the study of insect spatial distribution: practical implications in the integrated management of orchard and vineyard pests

“…This limitation may in part be overcome if efforts are directed to the development of intelligent Location-Aware Systems that allow automation of trapping devices and treatment operations (Wen et al 2009;Pontikakos et al 2012).…”

An insect classification analysis based on shape features using quality threshold ARTMAP and moment invariant

“…Wen, C. et al [54] also developed an automatic identification system that is capable of detecting the common pests' insects in orchards industry. This system utilizes the invariant region feature detector in order to extract vital features of the insect images.…”

“…These neural networks are Fuzzy ARTMAP (FAM) [6][7][8], Gaussian ARTMAP (GAM) [55,57], Symmetrical ARTMAP (SyAM) [4], Fully Self-Organizing ARTMAP (FSOAM) [5] and Bayesian ARTMAP (BaAM) [49]. Basically, these networks can be categorized as an incremental learning technique [4,6,54]. Andonie, R. [3] had stated that a network with incremental learning will have several criteria and there are: (1) it should be able to learn additional information from new data (2) it should not required access to the original data used to train the existing system (3) it should preserve previously acquired knowledge (4) it should be able to accommodate new data categories that may be introduced with new data.…”