Abstract-Determination of rice seed varieties is very important to ensure varietal purity in the production of high-quality seed. To date, manual seed inspection is carried out to separate foreign rice seed varieties in rice seed sample in the laboratory as there is lack of an automatic seed classification system. This paper describes a simple approach of using image processing technique and artificial neural network (ANN) to determine rice seed varieties based on extracted colour features of individual seed images. The experiment was conducted using 200 individual seed images of two Malaysian rice seed varieties namely MR 219 and MR 269. The acquired seed images were processed using a set of image processing procedure to enhance the image quality. Colour feature extraction was carried out to extract the red (R), green (G), blue (B), hue (H), saturation (S), value (V) and intensity (I) levels of the individual seed images. The classification using ANN was carried out by dividing the data sets into training (70% of data), validation (15%) and testing (15%) dataset respectively. The best ANN model to determine the rice seed varieties was developed, and the accuracy levels of the classification results were 67.5% and 76.7% for testing and training data sets using 40 hidden neurons.

Soil is essential for plant growth. The soil provides support for plant, medium for root growth, and most importantly it offers nutrients for plant uptake. The nutrients variability in soil is vary depending on several factors such as soil type, soil microbes, and soil pH. Therefore, nutrients available in the soil is important to be mapped to investigate the state of nutrient present. In this study, the available content of Nitrogen (N), Phosphorus (P) and Potassium (K) were determined for a high-density planting systems of Harumanis mango plants grown in the greenhouse. Thirty-two soil samples were collected from a greenhouse for analysis of NPK content in top soil. The soil was analyzed using the Kjeldahl method, UV Spectrophotometer and Atomic Absorption Spectroscopy (AAS) for N, P, and K content respectively. These amount of macronutrients were then mapped with respective georeferenced location to produce NPK nutrients maps using standard classification in ArcGIS software. Results have indicated that N, P, and K were ranged between 0.06-0.12% (very low), 4-648 ppm (low-very high) and 0.02-1.37 cmol/kg (low-high) accordingly. Overall, it can be concluded, soil in selected greenhouse is poor in N, high in P and moderate in K content. Hence, it is suggested more N and adequate amount of K fertilizer should be supplied to increase the plant's productivity. The produced maps give a new perspective in a farming management concept in term of variable rate fertilizer application for Harumanis mango plants grown in the greenhouse.

Aroma and other physicochemical parameters are important attributes influencing consumer perception and acceptance of rice. However, current methods using multiple instruments and laboratory analysis make these assessments costly and time-consuming. Therefore, this study aimed to assess rice quality traits of 17 commercial rice types using a low-cost electronic nose and portable near-infrared spectrometer coupled with machine learning (ML). Specifically, artificial neural networks (ANN) were used to classify the type of rice and predict rice quality traits (aromas, color, texture, and pH of cooked rice) as targets. The ML models developed showed that the chemometrics obtained from both sensor technologies successfully classified the rice (Model 1: 98.7%; Model 2: 98.6%) and predicted the peak area of aromas obtained by gas chromatography-mass spectroscopy found in raw (Model 3: R = 0.95; Model 6: R = 0.95) and cooked rice (Model 4: R = 0.98; Model 7: R = 0.96). Furthermore, a high R = 0.98 was obtained for Model 5 to estimate the color, texture, and pH of cooked rice. The proposed method is rapid, low-cost, reliable, and may help the rice industry increase high-quality rice production and accelerate the adoption of digital technologies and artificial intelligence to support the rice value chain.

No abstract