Philippine E-Journals

Optimizing Mushroom Farming: Fruiting Quality Prediction and Automation through Machine Learning

Jaeruz Timothy M Datiles | Mia A Fulgueras | Mark Daniel J Macapagong | Marvin B Montaña

Discipline: Computer Engineering

 

Abstract:

In the Philippines, most of mushroom cultivators practice traditional method of mushroom farming which is time consuming and labor intensive. The study entitled ‘OPTIMIZING MUSHROOM FARMING: FRUITING QUALITY PREDICTION AND AUTOMATION THROUGH MACHINE LEARNING’ - KABUTECH aim to keep the Oyster mushroom growing at the optimal temperature, relative humidity, carbon dioxide, and illuminance in a 450 x 330 x 300 mm monitored environment box. The application of Internet of Things (IoT) and Machine Learning help in reducing human labor in mushroom farming and also increase oyster mushroom productivity by controlling the most suitable environment for the mushroom to grow in. The proposed device employs supervised vector machine (svm), a set of supervised machine learning method predict the quality of the mushroom based on the parameters and IoT to monitor parameters status. The researchers used the incremental method. Optimizing mushroom farming involves many different process that can be done simultaneously and independently. In mushroom cultivation, monitoring is important to maintain the consistency of mushroom quality. The quality of the mushroom varies upon the changes occurring in its environment. Temperatures of 25°C to 30°C, relative humidity of 82% to 98%, CO2 of 600-800ppm, and illuminance of 50-1200 lux are ideal for oyster mushroom growth and development. The result of the software material and prototype evaluation from 10 experts and 50 normal users was rated very good and excellent, this simply means that the design proposal is practical and implementable.