Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail

Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail

Rubber breeding program is initiated by Rubber Research Institute of Malaysia for focusing on the best clone production which relies on the quality of the latex yielding for both latex and latex timber clones and known as RRIM clone series. Even though RRIM had overcome the problem to increase the y...

Full description

Saved in:
Bibliographic Details
Main Author: Ismail, Faridatul Aima
Format: Thesis
Language:English
Published: 2017
Subjects:
Rubber industry
Online Access:https://ir.uitm.edu.my/id/eprint/38016/1/38016.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my-uitm-ir.38016
record_format uketd_dc
spelling my-uitm-ir.380162023-01-26T08:59:22Z Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail 2017 Ismail, Faridatul Aima Rubber industry Rubber breeding program is initiated by Rubber Research Institute of Malaysia for focusing on the best clone production which relies on the quality of the latex yielding for both latex and latex timber clones and known as RRIM clone series. Even though RRIM had overcome the problem to increase the yield of latex, but the difficulties in differentiating the clone series still prevailing due to lack of information in reference books and required skill from the expertise. Therefore, the objectives are to design measure and discriminate the latex of RRIM clones using optical sensing indicator via reflectance technique. This indicator used Near-Infrared LED to transmit the light and the photodiode will received the reflectance light and converted into voltage. The measured voltage then analyzed using SPSS tool to investigate the discrimination between clones. Based on the analysis, there are three clones (RRIM2002, 2007 and 3001) are distinguished between each other with p-value 0.000. However RRIM2008 and 2014 not distinguished amongst others and could be seen as one group. Yet these two groups are discriminated between each other. By completing the analysis, only four clones were decided to use as an input parameter for ANN toolbox. The best optimized hidden layer size at ±0.5 fixed thresholds was 25 neurons with the testing accuracy of 74.4% and the true positive rate of this case is 73.9% (RRIM2002), 72.2% (RRIM2007), 61.4% (RRIM2008) and 94.1% (RRIM3001). Based on the findings, the objectives to measure latex via reflectance, identify and discriminate the selected RRIM clones series was successfully achieved. 2017 Thesis https://ir.uitm.edu.my/id/eprint/38016/ https://ir.uitm.edu.my/id/eprint/38016/1/38016.pdf text en public masters Universiti Teknologi MARA (UiTM) Faculty of Electrical Engineering Madzhi, Nina Korlina
institution Universiti Teknologi MARA
collection UiTM Institutional Repository
language English
advisor Madzhi, Nina Korlina
topic Rubber industry
spellingShingle Rubber industry
Ismail, Faridatul Aima
Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
description Rubber breeding program is initiated by Rubber Research Institute of Malaysia for focusing on the best clone production which relies on the quality of the latex yielding for both latex and latex timber clones and known as RRIM clone series. Even though RRIM had overcome the problem to increase the yield of latex, but the difficulties in differentiating the clone series still prevailing due to lack of information in reference books and required skill from the expertise. Therefore, the objectives are to design measure and discriminate the latex of RRIM clones using optical sensing indicator via reflectance technique. This indicator used Near-Infrared LED to transmit the light and the photodiode will received the reflectance light and converted into voltage. The measured voltage then analyzed using SPSS tool to investigate the discrimination between clones. Based on the analysis, there are three clones (RRIM2002, 2007 and 3001) are distinguished between each other with p-value 0.000. However RRIM2008 and 2014 not distinguished amongst others and could be seen as one group. Yet these two groups are discriminated between each other. By completing the analysis, only four clones were decided to use as an input parameter for ANN toolbox. The best optimized hidden layer size at ±0.5 fixed thresholds was 25 neurons with the testing accuracy of 74.4% and the true positive rate of this case is 73.9% (RRIM2002), 72.2% (RRIM2007), 61.4% (RRIM2008) and 94.1% (RRIM3001). Based on the findings, the objectives to measure latex via reflectance, identify and discriminate the selected RRIM clones series was successfully achieved.
format Thesis
qualification_level Master's degree
author Ismail, Faridatul Aima
author_facet Ismail, Faridatul Aima
author_sort Ismail, Faridatul Aima
title Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
title_short Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
title_full Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
title_fullStr Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
title_full_unstemmed Development of reflectance indicator using optical sensing on latex / Faridatul Aima Ismail
title_sort development of reflectance indicator using optical sensing on latex / faridatul aima ismail
granting_institution Universiti Teknologi MARA (UiTM)
granting_department Faculty of Electrical Engineering
publishDate 2017
url https://ir.uitm.edu.my/id/eprint/38016/1/38016.pdf
_version_ 1783734449431642112

Similar Items