Pest outbreaks are still threatening crop production particularly where manual field reports are infrequent. Manual scouting is slow, causes more losses to the crop and, in most cases, excess in the use of pesticides. This work introduces an independent pest-monitoring system, which is a pest-identification framework based on images, environmental sensing based on IoTs, and decision logic applied using AI. This system works with the help of lightweight deep-learning models that help detect pests in the field and utilizes distributed sensor nodes to detect temperature, humidity, and the state of soil. These two data streams can be used to automatically perform such actions as local spraying and automated irrigation. This integrated solution lowers the workforce, enhances accuracy in the detection, and helps in environmentally-friendly pest management.