Here is an article about a method that uses near-infrared images collected with drones. But smartphone cameras are sensitive to infrared. If you record a movie of a TV remote, the IR LED light is visible on the display. Maybe it yields some useful data if you partition images taken by smartphone into some squares and use the RGB histograms of them for pattern classification.
CropSpec sensors measure plant reflectance to determine chlorophyll content, which correlates to nitrogen concentration in the leaf. This non-destructive, non-contact measurement method provides accurate, stable readings with repeatable values. Plant nanobiosensors detect physiological signals such as presence of pathogen, temperature change, acidity, or volatile organic compounds, and communicate vital information regarding plant health. The received information can interpret the degree of damage or overall welfare of the plant. Optical sensors are also used to study the crop vigour by including the biomass of the soil and Nitrogen to other gases ratio in the soil as variables. This helps farmers regulate the moisture levels in the air and soil and prevent damp conditions. The soil moisture sensor is one of the most important agricultural sensors. Soil moisture determines the water supply status of crops. Too high or too low soil moisture will affect the normal growth of crops above the ground. Now, farmers in India are adopting smart methods of farming to save time, labour, cost, and money. The different types of agriculture sensors in use include soil moisture sensors, temperature & humidity sensors, and nutrient sensors. Having data is important to making accurate predictions and making informed decisions. Smart agriculture sensors provide farmers with that data, so they can make informed decisions about their farms, crops, fields, and equipment, and so they can plan for the future. The model predicts the crop yield for a specific crop. The model also recommends the most profitable crop and suggests the right time to use the fertilizers. The main objective is to obtain a better variety of crops that can be grown over the season. To predict the crop yield, selected Machine Learning algorithms such as Support Vector Machine (SVM), Artificial Neural Network (ANN), Random Forest (RF), Multivariate Linear Regression (MLR), and K-Nearest Neighbour (KNN) are used. Among them, the Random Forest showed the best results with 95% accuracy. Machine learning can also help farmers identify the most profitable crops to plant based on market demand and environmental factors. By analyzing historical market data and weather patterns, machine learning models can predict the demand for different crops and suggest optimal planting times and locations. Machine learning-based recommendation systems are powerful engines using machine learning (ML) algorithms to segment customers based on user data and behavioral patterns and target them with personalized product or content suggestions.