Reducing production cost and producing a better quality of crops have always been fundamental goals for the Florida citrus industry. Especially, it is applicable in recent years due to the widespread of an exotic disease, Huanglongbing (HLB). Precision agriculture technology is one of the solutions to the agricultural challenges by adopting the site-specific crop management of agricultural fields. With advanced sensing technologies, various field data can be collected to analyze in-field spatial variability of different cropping factors. This research focused on developing intelligent sensing systems to improve the capacity of current sensing technologies in agricultural applications and to support precision agriculture practices in the Florida citrus industry. In this research, four machine vision systems as nondestructive measurements were developed to monitor cropping factors at different crop growth stages. First, at the early fruit development stage, a site-specific yield forecast system was developed using a novel depth data transformation method that can identify the structure of three-dimensional fruit surface. Second, for pre-harvest to harvesting period, two machine vision systems to detect premature fruit drops were developed to measure the severity of fruit tree diseases such as the HLB. Finally, for the post-harvest stage, a citrus disease and defect inspection system was developed using a transfer learning technique and a real-time video processing algorithm with a graphical processing unit (GPU). The final accuracy of each system was 89.2%, 88.1%, 89.6%, and 94.9%, respectively. The intelligent sensing systems developed in this research can be used in commercial citrus groves to collect field data such as potential yields, the severity of disease, and the types of defects in the groves. The collected data can provide better management of citrus groves by helping the growers prepare tree inventory, fertilizers, and treatment of diseases. Future studies of this research will focus on combining the developed sensors with unmanned vehicles and robotics to accomplish the fully-automated crop production system and reduce humanity's footprint in agriculture.