Browsing by Author "Wijesinghe, A"

Now showing 1 - 3 of 3

Embargo
Digital Creation of Color Illusion Fabricated by Overlaying Different Colored Translucent Textiles Using Images
(IEEE, 2019-12-18) Wijesinghe, A; Seneviratne, L; Abeyratne, S
Overlaying different colored textiles which are translucent is a straight forward task to complete physically, in contrast this task is difficult to achieve digitally. Amount of information obtained from an image is limited, which is a major difficulty faced when using images to identify the features of a textile such as color, material, texture, thickness and transparency. An algorithmic approach is taken based on three hypotheses; random superimposing, background replacement and color augmentation. These techniques are based on; color identification, background replacement, random selection, pixel superimposing, color blending and image color augmenting. The algorithms are researched, implemented, experimented in-depth and critically compared. Four algorithms are implemented, two based on randomly superimposing and one each based on background replacement and color augmentation. Background replacement algorithm was hardly able to complete the task effectively, thus is the lowest ranked algorithm. In contrast, randomly superimposing and color augmenting algorithms were capable of carrying out the task successfully. Randomly superimposing costed the least time to complete, but the generated images were unnatural whereas color augmenting produced a perfectly natural image though the color of the final output was inaccurate. Further refining the color prediction algorithm is proposed to develop a more effective system.
Embargo
Machine Learning Based Automated Speech Dialog Analysis Of Autistic Children
(IEEE, 2019-10-24) Wijesinghe, A; Samarasinghe, P; Seneviratne, S; Yogarajah, P; Pulasinghe, K
Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic children at early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection and Vocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.
Embargo
Machine learning based automated speech dialog analysis of autistic children
(IEEE, 2019-10-24) Wijesinghe, A; Samarasinghe, P; Seneviratne, S; Yogarajah, P; Pulasinghe, K
Children with autism spectrum disorder (ASD) have altered behaviors in communication, social interaction, and activity, out of which communication has been the most prominent disorder among many. Despite the recent technological advances, limited attention has been given to screening and diagnosing ASD by identifying the speech deficiencies (SD) of autistic children at early stages. This research focuses on bridging the gap in ASD screening by developing an automated system to distinguish autistic traits through speech analysis. Data was collected from 40 participants for the initial analysis and recordings were obtained from 17 participants. We considered a three-stage processing system; first stage utilizes thresholding for silence detection and Vocal Activity Detection for vocal isolation, second stage adopts machine learning technique neural network with frequency domain representations in developing a reliant utterance classifier for the isolated vocals and stage three also adopts machine learning technique neural network in recognizing autistic traits in speech patterns of the classified utterances. The results are promising in identifying SD of autistic children with the utterance classifier having 78% accuracy and pattern recognition 72% accuracy.