Abstract:
Cloves are the greatest valuables spices which have the common uses for food preservatives and other many therapeutics purposes like medicines, toothpaste, skin creams and soaps. Cloves originated from Indonesia and has been cultured in different part of the words, such as Zanzibar, Pakistan and so on. Cloves are sources of income, thus improving production they need best evaluation, processing and monitoring which will potentially contribute to national and international economy, therefore, to archives this it needs strong commandment on evaluation processing. The main problem faced in cloves quality evaluation to position an International standard is the lack of technology that will automatically evaluate the quality of the cloves instead of using traditionally, trial and error evaluation practices of physical touching and eye contacts. The major target of the present research is to design and develop low cost system that will automatically detect, evaluate and give out the results of the quality cloves based on different qualities, such as first, second and third qualities, aiming to catalogue valuable information of the process for cloves quality evaluation improvement. The system will determine and evaluate the color frequencies, presences of water vapour and temperature on the dried cloves by using TCE3200 and DHT22 sensors, then analyses it to provide the results superiority of the cloves. The system will use Arduino UNO microcontroller version AT mega328P in order to process input, display output via LCD and finally through GSM SIM800L module send the data to the cloud platform (ThingSpeak) for visualization and analysis. An Integrated Development Software Environment (IDE) of Arduino as an open source application will use for code generation and then upload it to the Microcontroller. The machine-learning principles thorough Fuzzy Inferences System (FIS) will use in prediction analysis for quality of cloves. It is a ruled-based expert system used to provide the output according to the inputs three sated parameters of dry cloves such as intensity as a color reading of dry cloves, temperature and moisture levels. In the end the quality will be determine based on three main samples (A, B and C), A, represent for the first quality, B for the second quality and C for the third quality. On each, a sensor is able to read five times different types quality of the cloves in order to obtain mean, standardization, regression and core relation coefficient between them. The sensors are trained to read color frequencies, temperature and moisture of dry cloves and after that, the data is drive to cloud for analytics and visualization. Due to the limitations of the ThingSpeak platform in doing data analysis and visualization, MATLAB application also used for data clearance.
