Open Digital Repository Lanka Education & Research Network
Not a member yet
28 research outputs found
Sort by
Detection of White Leaf Disease in Sugarcane Crops Using UAV-Derived RGB Imagery with Existing Deep Learning Models - Kandy
No full textWhite leaf disease (WLD) is an economically significant disease in the 444444sugarcane industry. This work applied remote sensing techniques based on unmanned aerial vehicles (UAVs) and deep learning (DL) to detect WLD in sugarcane fields at the Gal-Oya Plantation, Sri Lanka. The established methodology to detect WLD consists of UAV red, green, and blue (RGB) image acquisition, the pre-processing of the dataset, labelling, DL model tuning, and prediction. This study evaluated the performance of the existing DL models such as YOLOv5, YOLOR, DETR, and Faster R-CNN to recognize WLD in sugarcane crops. The experimental results indicate that the YOLOv5 network outperformed the other selected models, achieving a precision, recall, mean average [email protected] ([email protected]), and mean average [email protected] ([email protected]) metrics of 95%, 92%, 93%, and 79%, respectively. In contrast, DETR exhibited the weakest detection performance, achieving metrics values of 77%, 69%, 77%, and 41% for precision, recall, [email protected], and [email protected], respectively. YOLOv5 is selected as the recommended architecture to detect WLD using the UAV data not only because of its performance, but this was also determined because of its size (14 MB), which was the smallest one among the selected models. The proposed methodology provides technical guidelines to researchers and farmers for conduct the accurate detection and treatment of WLD in the sugarcane fields. (TEST EDIT?
Prof
No full textAbstract
Background
Although the assessment of disease burden should be a priority for allocating resources, leptospirosis is grossly underestimated despite its true burden in Sri Lanka. This study aimed to assess the morbidity and mortality of leptospirosis based on routine surveillance data, hospital reported data and scientific publications from Sri Lanka.
Method
A systematic review was carried out, and Pub Med, MEDLINE®, BIOSIS Previews, Zoological Record, Web of Science Core Collection, Current Contents Connect, KCI-Korean Journal Database, BIOSIS Citation Index, Data Citation Index, SciELO Citation Index and Google Scholar databases were searched. Quarterly epidemiological bulletin (QEB), indoor morbidity & mortality returns (IMMR) and hand searches of local literature were performed in local libraries. Forty-two relevant full texts, 32 QEBs, and 8 IMMR were included in the full text review. Adjustments were made for under diagnosis, underreporting and chance variability.
Results
The estimated annual caseload of leptospirosis in Sri Lanka from 2008 to 2015, was 10,423, and the cumulative annual incidence of leptospirosis that required hospitalization was 52.1 (95% CI 51.7–52.6) per 100,000 people. The estimated number of annual deaths due to leptospirosis was approximately 730 (95% CI 542–980), with an estimated pooled case fatality ratio of 7.0% (95% CI 5.2–9.4). The most common organs involved were the kidney, liver and heart, with median rates of 48.7, 30, and 14.2%, respectively.
Conclusion
Our systematic review shows gross underestimation of the true leptospirosis burden in the national statistics of Sri Lanka, and the hospitalization rates estimated in our study were compatible with the total burden estimate of 300·6 (95% CI 96·54–604·23) per 100,000 people published previously
Science and bioethics of CRISPR-Cas9 gene editing: an analysis towards separating facts and fiction
No full textgene editing and it's implicationsSince its emergence in 2012, the genome editing technique known as CRISPR-Cas9 and its scientific use have rapidly expanded globally within a very short period of time. The technique consists of using an RNA guide molecule to bind to complementary DNA sequences, which simultaneously recruits the endonuclease Cas9 to introduce double-stranded breaks in the target DNA. The resulting double-stranded break is then repaired, allowing modification or removal of specific DNA bases. The technique has gained momentum in the laboratory because it is cheap, quick, and easy to use. Moreover, it is also being applied in vivo to generate more complex animal model systems. Such use of genome editing has proven to be highly effective and warrants a potential therapy for both genetic and non-genetic diseases. Although genome editing has the potential to be a transformative therapy for patients it is still in its infancy. Consequently, the legal and ethical frameworks are yet to be fully discussed and will be an increasingly important topic as the technology moves towards more contentious issues such as modification of the germline. Here, we review a number of scientific and ethical issues which may potentially influence the development of both the technology and its use in the clinical setting.no sponso
Image Classification of Paddy Field Insect Pests Using Gradient-Based Features
No full textAgriculture is one of the principal economic activities of the Jaffna peninsula in the Northern Province of Sri Lanka. Over 60% of the work force in the peninsula depends on agriculture for their livelihood. Paddy cultivation in the peninsula contributes substantially to the gross national income of the country. Such Paddy crops are affected by the attack of insect pests. Therefore paddy field insect pest identification is an important task to the sustainable agricultural development in the Jaffna peninsula. This paper offers a framework to classify images of paddy field insect pests using gradient-based features through the bag-of-words approach. Images of twenty classes of paddy field insect pests were obtained from Google Images and photographs taken by the Faculty of Agriculture, University of Jaffna, Sri Lanka. The images were then classified through the system that involves identification of regions of interest and representation of those regions as scale-invariant feature transform (SIFT) or speeded-up robust features (SURF) descriptors, construction of codebooks which provides a way to map the descriptors into a fixed-length vector in histogram space, and the multi-class classification of the feature histograms using support vector machines (SVMs). Furthermore, the
histograms of oriented gradient (HOG) descriptors were applied in classification. As a baseline classifier the nearest neighbour approach was used and compared with SVM-based classifiers. Testing results show that HOG descriptors significantly outperform existing local-invariant features: SIFT and SURF in paddy field insect pests classification. HOG descriptors when combined with SURF features yield around 90% accuracy in classification. For simplicity and speed, linear SVM was used as a classifier throughout the study