No abstract

The standard practice among rice farmers in Malaysia is to apply fertilizer using a single application rate for the whole field. However, fertility conditions vary across the field. The excess use of fertilizer leads to increased input cost and can be damaging to the environment. The focus of this research was to develop a method to apply fertilizer on-the-go while sensing the crop nutrient status of rice plants. A machine learning approach was used to develop a crop nitrogen status prediction model. The model used spectral data from an active canopy reflectance sensor and several vegetation indices as inputs. The model was then incorporated into an on-the-go variable rate fertilizer application system. System performance was then evaluated in the field. The results from this work showed that the model had and accuracy of 83% in classifying the nitrogen status of the rice plants. The results also showed that our method was able to save up to 20% fertilizer use while maintaining yield. These findings are important for large estate farmers who are looking to increase productivity and efficiency.

This work reviews the current state of the art for pineapple production in Malaysia from the perspective of mechanization and automation. It examines the issues and challenges facing this industry. The review has led us to the conclusion that pineapple production still relies heavily on manual labour. The problems facing this industry is no different than other food crops in that low yield labour and high cost are the primary issues that need to be tackled. Although numerous engineering research work to overcome production issues has been done for crops such as rice and maize, engineering research for pineapples has been scarce. The lack of engineering research literature on this crop presents an opportunity for the scientific community to invest effort in this relatively untapped industry. This work further proposes areas where the use of Industry 4.0 technologies can be exploited in order to increase productivity and reduce input costs. Cyber-physical systems that could address issues in planting, crop maintenance and harvesting are put forth as a possible solution.

Rice serves as the primary food source for nearly half of the global population, with Asia accounting for approximately 90% of rice production worldwide. However, rice farming faces significant losses due to pest attacks. To prevent pest infestations, it is crucial to apply appropriate pesticides specific to the type of pest in the field. Traditionally, pest identification and counting have been performed manually using sticky light traps, but this process is time-consuming. In this study, a machine vision system was developed using a dataset of 7328 high-density images (1229 pixels per centimetre) of planthoppers collected in the field using sticky light traps. The dataset included four planthopper classes: brown planthopper (BPH), green leafhopper (GLH), white-backed planthopper (WBPH), and zigzag leafhopper (ZIGZAG). Five deep CNN models—ResNet-50, ResNet-101, ResNet-152, VGG-16, and VGG-19—were applied and tuned to classify the planthopper species. The experimental results indicated that the ResNet-50 model performed the best overall, achieving average values of 97.28% for accuracy, 92.05% for precision, 94.47% for recall, and 93.07% for the F1-score. In conclusion, this study successfully classified planthopper classes with excellent performance by utilising deep CNN architectures on a high-density image dataset. This capability has the potential to serve as a tool for classifying and counting planthopper samples collected using light traps.

No abstract