Browsing by Author "Dangalle, C. D"

Now showing 1 - 4 of 4

Open Access
Arboreal Tiger Beetles Recorded from Lowland Crop Cultivations in Sri Lanka
(UOC e-Repository > Science Department of Zoology, 2021-01) Abeywardhana, D. L; Dangalle, C. D; Mallawarachchi, Y
Purpose : Thirty-one species of arboreal tiger beetles are known from Sri Lanka of which 25 species are endemic. However, their habitat types are poorly documented and the available records are far outdated. Therefore, a survey of tiger beetles was carried out to determine their present occurrence with emphasis on agricultural habitat types. Research Method : Forty-six locations of the country, covering eighteen districts, all provinces, representing a majority of bioclimatic zones except those in Montane Sri Lanka were surveyed for arboreal tiger beetles. Sampling was conducted using the visual encounter method. Collected beetles were identified using taxonomic keys and descriptions. Findings : Eight species of arboreal tiger beetles were collected from the survey. Majority of the species (06) were collected from crop cultivations of coconut and also from tea, fruit farms, betel leaf, cinnamon and pepper. Four species of Derocrania and two species of Tricondyla were recorded from the cultivations and all had fused elytra and hence unable to fly. Derocrania scitiscabra was the dominant arboreal tiger beetle species in the crop cultivations. Originality/ Value : The study documents hitherto unrecorded habitat types for a poorly documented important beetle group of Sri Lanka. It further provides information for future research on the possibility of using arboreal tiger beetles as bio-control agents of insect pests of agricultural crops.
Open Access
Cicindelinae of Sri Lanka: New record of the arboreal tiger beetle Tricondyla gounellei Horn, 1900
(NSF, 2020-05-29) Abeywardhana, D. L; Dangalle, C. D; Mallawarachchi, Y
Information is provided on the newly recorded Tricondyla gounellei Horn, 1900, an arboreal tiger beetle, hitherto known only from Southern India, with this being its fi rst from Sri Lanka. Following fi eld surveys conducted from 2017 to 2019 in forty-one locations in the country, this species was recorded from two locations namely, Vellankulam in Mannar District and Kirinda in Hambantota District. Tricondyla gounellei, closely resembles Tricondyla granulifera Motschulsky, 1857 previously recorded from Sri Lanka. However, T. gounellei can be distinguished from T. granulifera by the smaller body size, short elytra that are narrower in the middle and palpi with black terminal joints which in T. granulifera is red.
Embargo
Deep learning approach to classify Tiger beetles of Sri Lanka
(Elsevier, 2021-05-01) Abeywardhana, D. L; Dangalle, C. D; Nugaliyadde, A; Mallawarachchi, Y
Deep learning has shown to achieve dramatic results in image classification tasks. However, deep learning models require large amounts of data to train. Most of the real-world datasets, generally insect classification data does not have large number of training dataset. These images have a large amount of noise and various differences. The paper proposes a novel architectural model which removes the background noise and classify the Tiger beetles. Here object location is identified using contours by converting the original coloured image to white on black background. Then the remaining background is eliminated using grabcut algorithm. Later the extracted images are classified using a modified SqueezeNet transfer learning model to identify the tiger beetle class up to genus level. Transfer learning models with fewer trainable parameters performed well than the total number of parameters in the original model. When evaluating results it was identified that by freezing uppermost layers of SqueezeNet model better accuracy can be gained while freezing lowermost layers will reduce the validation accuracy. The proposed model achieved more than 90% for the test set in 40 epochs using 701,481 trainable parameters by freezing the top 19 layers of the original model. Improving the pre-processing to localize insect has improved the accuracy.
Open Access
An ultra-specific image dataset for automated insect identification
(Springer US, 2022-01-09) Abeywardhana, D. L; Dangalle, C. D; Nugaliyadde, A; Mallawarachchi, Y
Automated identifcation of insects is a tough task where many challenges like data limitation, imbalanced data count, and background noise needs to be overcome for better performance. This paper describes such an image dataset which consists of a limited, imbalanced number of images regarding six genera of subfamily Cicindelinae (tiger beetles) of order Coleoptera. The diversity of image collection is at a high level as the images were taken from diferent sources, angles and on diferent scales. Thus, the salient regions of the images have a large variation. Therefore, one of the main intentions in this process was to get an idea about the image dataset while comparing diferent unique patterns and features in images. The dataset was evaluated on diferent classifcation algorithms including deep learning models based on diferent approaches to provide a benchmark. The dynamic nature of the dataset poses a challenge to the image classifcation algorithms. However transfer learning models using softmax classifer performed well on the current dataset. The tiger beetle classifcation can be challenging even to a trained human eye, therefore, this dataset opens a new avenue for the classifcation algorithms to develop, to identify features which human eyes have not identifed.