Research Papers - Dept of Information Technology

Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/593

Browse

Filters

2023 - 20262

Advanced Search

Filter by

Settings

Subject: search.filters.subject.Computer vision

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    PublicationOpen Access
    Vision incorporated MUltichannel Feature Fusion Template Matching (MUFF-TM) and real-time sub-pixel coordinate localization for 2D textile surface in ultrasonic tacking systems
    (Elsevier B.V., 2026-09) Kahandawala, B, S; Nalmi, R; Sodige, B, A.K; Subasinghage, K; Silva, B, N; Wijesinghe, R,E; Woo, S, T
    Temporary stitches are essential in apparel manufacturing as they temporarily secure fabric pieces to prevent misalignment during machine sewing and ensure high-quality results. Manual ultrasonic tacking machines were introduced to enhance the precision of this process; however, the necessity for expert operators remains a major constraint. This work introduces a real-time system for ultrasonic tacking machines that combines vision-guided single-modal MUltichannel Feature Fusion Template Matching (MUFF-TM) to autonomously identify and align tacking points on textiles with sub-pixel spatial accuracy. To overcome the limitations of classic feature-based algorithms on smooth and deformable fabrics, the proposed method utilizes macro-contour extraction and equidistant boundary sampling rather than relying on unstable local textures. Experimental results demonstrate that MUFF-TM achieves a 100% target detection rate with a highly stable Mean Absolute Error (MAE) of under 5 pixels across various dynamic conditions, including changes in orientation, illumination, scale, and non-rigid deformation. By significantly outperforming traditional algorithms (SIFT, SURF, and ORB) in spatial precision, the developed software interface offers a robust, versatile, and scalable solution for advancing automated precision in the apparel industry. Copyright
  • Thumbnail Image
    PublicationEmbargo
    Development of an Elephant Detection and Repellent System based on Efficient Det-Lite Models
    (IEEE, 2023-04-03) Pemasinghe, S; Abeygunawardhana, P.K.W
    Human-elephant conflict (HEC) has become a major concern in Sri Lanka that results in many unfortunate human and elephant deaths. Methods that are currently in place to mitigate HEC, such as electrical fences have undesirable consequences resulting in both human and elephant casualties. In this paper, we have proposed a method based on computer vision and deep learning that has a promising potential for detecting and repelling elephants without endangering the lives of elephants or humans. We have used EfficientDet-Lite models that provide a good compromise between accuracy and performance in order to be usable with a resource-constrained device like a Raspberry Pi.