Science
Permanent URI for this communityhttp://repository.kln.ac.lk/handle/123456789/1
Browse
3 results
Search Results
Item Off-line signature verification system using artificial neural networks(Research Symposium on Pure and Applied Sciences, 2018 Faculty of Science, University of Kelaniya, Sri Lanka, 2018) Ambegoda, A. L. T. P.; Perera, B. B. U. P.Handwritten signature recognition method is the most popular recognition method of a personal identity. But it is easy to misuse that the signature forgery has become a great threat to the accuracy of the documentary. In this paper, we present an off-line signature recognition method using an Artificial Neural Network (ANN) created using Matlab (Matrix Laboratory). A signature dataset consisting of 248 signatures (both genuine and forged) of three different owners, used to train the network. First, the signatures were preprocessed enabling extracting their features. Then some geometrical features were extracted from each signature to feed as the inputs to the neural network. Each image was converted to a binary image and after identifying the geometric center, the image was divided into four segments. Again identifying the geometric center of each segment, each segment was divided again into four segments. Geometrical features from each segment were extracted and used as inputs to feed the network. A single neural network was created to execute both authentication and verification steps of the signature recognition. The network topology is optimized to the given dataset. Supplying corresponding target values, the network was trained. The Mean Squared Error (MSE) function was used to determine the performance of the network. Changing the parameters, the network was trained until it gets a favorable output. It showed a favorable performance value of 0.164 and an accuracy greater than 72% for all three subsets: training, validation and test sets in the training dataset. Then the network was tested on an untrained dataset of the same owners. For this untrained dataset, a favorable result was gained with an accuracy of 0.6129 and performance value of 0.3226. The lower the performance value, the better the network. It is assumed that high variability, too simplicity, and consistency of the data affected the results of the network. It is proposed to consider a larger dataset and improve the algorithm to be more sensitive to the above mentioned misleading factors. Although the performance is greater for the testing dataset than for the training dataset, it is concluded that the created network can be enhanced and developed to be applied in practical situations.Item Effects of abuse of bhang on autonomic nervous system and cardiac electrophysiology: a study of Indian farmers.(International Research Symposium on Pure and Applied Sciences, 2017 Faculty of Science, University of Kelaniya, Sri Lanka., 2017) Nayak, S.K.; Pal, K.Cannabis sativa products are widely abused across the globe, just next to the abuse of alcohol because of their exultant impact. In India, the abuse of cannabis products is very much prevalent. This can be attributed to the fact that the price of making of the cannabis products is very low. The plant is cultivated comfortably in the Indian environment. Also, the cannabis products have been reported to be associated with a very low level of toxicity. Although the Government of India (GoI) has imposed restrictions on the abuse of the cannabis products, their abuse is widespread among the ordinary people. However, the restrictions imposed on the abuse of bhang (a product made from the leaves of cannabis plant) are not much strong. The people following Hindu religion also abuse bhang because of their religious perception. Bhang is employed by them to make drinks in some carnivals. Recently, several studies have reported that the abuse of cannabis products may lead to many cardiovascular as well as non-cardiovascular diseases and even death. An in-depth literature survey on the effect of cannabis products on human health suggested that only few studies have been performed to divulge information about the effect of cannabis products on physiology of the heart and the ANS. In the last few decades, the analysis of electrocardiogram (ECG) signal has received a special attention of the researchers. This is because it provides information about the cardiac electrophysiology. Heart rate variability (HRV) analysis provides a non-invasive method to analyze the physiology of the ANS. Taking a note of the afore-mentioned facts, a study was performed on Indian paddy field workers (volunteers) to understand the effect of bhang abuse on ANS and cardiac activity by acquiring their ECG signals for 5 min. The RR intervals (RRIs) were extracted from the ECG signals, and the time-domain, the frequency-domain, and the non-linear HRV features were computed from the RRIs. An in-depth analysis of the HRV features suggested that the abuse of bhang has increased the sympathetic activity in the volunteers abusing bhang regularly. Alteration in the cardiac electrophysiology was examined from the wavelet-based analysis of the ECG signals. The Daubechies (db06) wavelet was used to decompose the ECG signals into level 8, and the signal reconstruction was performed using D7+D8 sub-bands. 12 statistical features were extracted from the reconstructed signals and the statistical significance of the extracted features was examined using linear and non-linear statistical methods. Some of the features were found to be significantly different among the population abusing bhang and the population not abusing bhang, suggesting an alteration in the cardiac electrophysiology due to the abuse of bhang. Artificial neural network (ANN) was able to classify the HRV and the ECG data with an accuracy ≥90%.Item Patterns of species composition of beach seine fisheries off North-Western coast of Sri Lanka, fishers’ perceptions and implications for co-management,(Pergamon, 2016) Gunawardena, N.D.P.; Jutagate, T.; Amarasingha, U.S.As in many developing countries, small-scale fisheries including beach seining contribute significantly livelihoods and food security of coastal communities. Beach seining in Sri Lanka is seasonal mainly during calm season deprived of strong monsoonal winds, and essentially a multi-species fishery. Knowledge about the seasonal occurrence of pelagic species is important to be known for proper planning of the fishing activity, especially due to the reason that beach seine fishers in many parts of Sri Lanka make decisions to attach the cod-end of correct type depending on the target species. The possibility of identifying pattern of seasonal occurrence of target fish species in beach seine fishing sites off the southern region of north-western coast of Sri Lanka was therefore investigated using Self Organizing Maps (SOM). The analysis indicated that beach seine fishers’ local knowledge to predict the occurrence of certain species in the fishing sites to adjust their fishing strategies to target desirable species was consistent with the findings of SOM approach. Consequently, it was concluded that as beach seine fishers use indirect indicators such as colour of sea water and behaviour sea birds predict the species occurrence fairly accurately, their local knowledge can be incorporated in the management planning of beach seine fisheries in the North Western coastal area of Sri Lanka.