Browsing by Author "Ranasinghe, K"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    item: Article-Full-text
    Combined static and motion features for deep-networks-based activity recognition in videos
    (IEEE, 2019) Ramasinghe, S; Rajasegaran, J; Jayasundara, V; Ranasinghe, K; Rodrigo, R; Pasqual, AA
    Activity recognition in videos in a deep-learning setting—or otherwise—uses both static and pre-computed motion components. The method of combining the two components, whilst keeping the burden on the deep network less, still remains uninvestigated. Moreover, it is not clear what the level of contribution of individual components is, and how to control the contribution. In this work, we use a combination of CNNgenerated static features and motion features in the form of motion tubes. We propose three schemas for combining static and motion components: based on a variance ratio, principal components, and Cholesky decomposition. The Cholesky decomposition based method allows the control of contributions. The ratio given by variance analysis of static and motion features match well with the experimental optimal ratio used in the Cholesky decomposition based method. The resulting activity recognition system is better or on par with existing state-of-theart when tested with three popular datasets. The findings also enable us to characterize a dataset with respect to its richness in motion information.
  • Thumbnail Image
    item: Article-Abstract
    Gas transport parameters for compacted reddish - brown soil in Sri Lankan landfill final cover
    Wickramarachchi, P; Ranasinghe, K; Hamamoto, S; Kawamoto, K; Nawagamuwa, UP; Moldrup, P; Komatsu, T
    Gas exchange through (he compacted final cover soil at landfill sites plays a vital role for emission, fate, and transport of toxic landfill f,ases. This study involved measuring the soil-gas diffusivity tDp/Du, the ratio of gas diffusion coefficients in soil and free air) and air permeability (k,J for differently compacted soil samples (reddish-brown soil) from the final cover at the Maharagama landfill in Sri Lanka. The samples were prepared by either standard Proctor compaction or hand compaction to dry bulk densities of 1.60-1.94 g cm='. Existing and modified models for predicting Dp/ Do and k" were tested against the measured data. The simple, single-parameter Buckingham model predicted measured Dp/ Do values across compaction levels equally well or better than a dry bulk density (DBD) dependent model and a soilwater retention (SWR) dependent model. The measured ku values for differently compacted samples were highly affected by the compactionlevel and the sample moisture preparation method. Also, for air permeability, a single-parameter Buckingham-type k" model was most accurate in predicting k" in the differently compacted soil samples. Equivalent air-filled pore diameters