← All proceedings

146TH THE IIER INTERNATIONAL CONFERENCE

Proceedings cover: 146TH THE IIER INTERNATIONAL CONFERENCE

Hong Kong 27-12-2017

6 papers

  1. Production of Bioplastic from Banana Peels

    M.R.Gaonkar, Prashant Palaskar, Rishikesh Navandar

    Paper ID II-CBEHNGK-27127-7533 · Pages 1-3

    Download PDF
  2. “Soil Organic Carbon and Total Nitrogen in Himalayan Rangeland”

    Dilkumar Limbu, Madan Koirala, Zhanhuan Shang

    Paper ID II-NSEHGKG-27127-9165 · Pages 4-8

    Abstract

    The objective of this study is to identify and quantify the present status of carbon and nitrogen pool in Himalayan rangeland and to make recommendations for enhancing carbon and nitrogen storage for rangeland management. To meet the aforementioned objectives, the field study was conducted in 2011 -2013. The study showed that soil organic carbon was highest in legume seeding sub-plot in top soil (28.53 ± 2.6) t/ha of heavily grazed area. Similarly, total nitrogen was highest in bottom soil (2.81 ± 0.16) t/ha in legume seeding sub-plot of enclosed un-grazed area. Usually, heavily grazed and legume seeding sub-plots had more soil organic carbon and soil organic nitrogen concentration compared to others. The value of above ground biomass was in increasing trend with decreasing grazing intensity but for below ground biomass, it was just the reverse. On the basis of the results of this study, the grazing intensity is positively correlated with above ground and below ground biomass and soil organic carbon but no any response with soil total nitrogen and soil bulk density.

    Key words - Biomass, Carbon, Climate change, Himalayan, Soil, Rangeland

    Download PDF
  3. Context-Awareideal Ratio Mask Estimation for Speech Denoising Using Deep Neural Networks

    Hannah Lee, Jaegyu Choi, Deokgyu Yun, Seung Ho Choi

    Paper ID II-RIETHKGK-27127-8702 · Pages 9-10

    Abstract

    In this paper, we propose a context-aware approachfor speech denoising based on deep learning. In most studies, single deep neural network(DNN) is trained in various noisy environments. However, large variations of noisy environments can cause performance degradation. We develop environmentally-optimized models trained in each noisy environment, which is effective in restoring clean speech spectra under different noisy conditions.In this research work, we first classify noisy spectra through DNN models trained for classifying noise environment before noisy speech spectra are processed. We apply the proposed method to the estimation of ideal ratio mask (IRM) that is used for denoising in spectral domain. Experimental results showed that the proposed context-aware method performs better than that using non-context-awareness model.

    Keywords - Speech denoising, Context-awareness,Deep learning, Deep neural network,Ideal ratio mask

    Download PDF
  4. A Signal Separation Method Based on Sparsity Estimation of Source Signals and Non-Negative Matrix Factorization

    Siyeon Nam, Serin Hong, Deokgyu Yun, Seung Ho Choi

    Paper ID II-RIETHKGK-27127-8704 · Pages 11-12

    Abstract

    In order to improve the signal separation performance of the non-negative matrix factorization (NMF), sparse non-negative matrix factorization (Sparse NMF, SNMF) was developed. Existing SNMF algorithm uses arbitrarily determined sparseness without considering the sparseness of individual sound sources. In this paper, we propose a new signal separation method that estimates the sparseness according to the characteristics of a sound source and applies it to SNMF algorithm. Experimental results show that the proposed method has better performance than the existing NMF and SNMF.

    Keywords - Signal separation, Non-negative matrix factorization, Sparseness, Denoising.

    Download PDF
  5. A Frequency Response Normalization Method Based on Deep Neural Networksfor IOT Devices

    Deokgyu Yun, Jaegyu Choi, Hanna Lee, Seung Ho Choi

    Paper ID II-RIETHKGK-27127-8705 · Pages 13-14

    Abstract

    This paper presents a novel frequency response normalization method for IoT(Internet of Things) devices, which is based on deep neural networks. When connecting the sounds acquired by different smart phones, unnatural soundsare generated, which is mainly due to different frequency responses for each smart phone. To solve this problem, we need to normalize the frequency responses of smart phones. In this study, we propose a normalization method using deep neural network. The input of the neural network is the spectrum of the smart phone sound to be processed and the output is the ratio of input to the spectrum of the reference smart phone sound. Experimental results show that when the acoustic signals acquired by different smart phones are connected, the naturalness of the sound is improved through objective and subjective evaluation.

    Keywords - Frequency normalization, IoT(Internet of Things), Virtual reality, User created contents, Deep neural network.

    Download PDF
  6. In-Depth Investigation of Safety Management System in Hong Kong Construction Industry – Literature Review and Research Agenda

    Nicole s.n. Yiua, Daniel W.M. Chan A

    Paper ID II-SIMHNGK-27127-001 · Pages 15-19

    Abstract

    The construction industry is one of the major contributors to industrial fatalities. The safety management system has been introduced in 1980s to proactively tackle risks in construction work activities, so as to promote healthy and safe working environment to all employees on construction sites. However, there has been so far no comprehensive review on the effectiveness of the implementation, except the accident rate. Against this background, the reliability, efficiency and effectiveness of the safety management system need to be evaluated and investigated. This study would carry out an in-depth investigation into the safety management system for the Hong Kong construction industry. The research results would generate some valuable insights into the development of corresponding effective strategies for enhancing the prevailing safety management system and facilitate a zero harm working environment in Hong Kong construction industry.

    Keywords - Safety management system, safety performance, safety characteristics, construction industry

    Download PDF