Muhammad Hafiz Bin Md Zaini
Thesis
In the year 2003, statistic shows that 97% of Malaysian eat cooked rice twice daily which mean by the average of two and half plate a day. Little that they know how hard it is for the farmers as there is a famous proverb in Malay which says “sebiji beras ibarat setitk peluh petani”, literally it means that the farmers has to work very hard in order to produce such high quality products where everyone can enjoy them during breakfast for “nasi lemak”, lunch for “nasi kandar” and dinner for anything that can include rice. Currently almost paddy farmer in Malaysia has little knowledge and technologies in detecting paddy disease while it is still in their early stages thus making the production of rice reduce by each year. One of the most common way for the farmer to detect paddy disease is by using eyesight as most paddy disease has its own unique distinction of features to detect it. This shows that how important early disease detection is as if any of the disease is detected the farmers can make further action to curb it from become even worst. An alternative to detect paddy disease can be implemented by using one of the Artificial Intelligence technique. This project will be focus on the application of real time image processing along with the usage of Back Propagation Neural Network (BPNN) to identify whether the paddy plant is healthy or not. By using real time image processing it will pre-process the image of the paddy leaf in real time and the result of it will be used as the input for the BPNN to learn the distinguish patterns of both healthy and infected leaf. With enough data this project could gain a significant results but it still comes out with a flaw where it could not detect properly whether the leaf is infected as some of the disease holds almost identical pattern thus making it hard for the system to identify it.
Paddy, Disease, Neural Network, and Real-time
Published on 01/02/2015
Submitted on 14/07/2016 01:26:20 am
UiTM Shah Alam
+1111