In this work, we present a novel approach for the early detection and diagnosis of skin diseases in farm animals, a major concern that can lead to reduced productivity, decreased animal welfare, and economic losses. Using Internet of Things (IoT) and MobileNetV2, we have developed a system that is built using Raspberry Pi for the gateway and low-power ESP 32 microcontrollers for sensor attachment. This system consists of sensors placed on the animals" bodies, including an electrocardiogram (ECG) sensor and a DS18B20 temperature sensor, which continuously monitor the animals" vital signs and skin temperature. The collected data is transmitted to a central server where it is processed using MobileNetV2, a deep learning model trained to recognize three common skin diseases in farm animals: Dermatophilosis, Dermatophycosis, and Papillomatosis. The results of this processing are then made available to animal owners and farmers through a mobile app. Our results show that the proposed system can accurately detect and diagnose skin diseases in farm animals with a high degree of recall (0.96), precision (0.96), and f1 score (0.96). The use of IoT and machine learning allows for realtime monitoring and early detection of skin diseases, which can significantly reduce the spread of infection and improve the overall health and welfare of farm animals. In addition, the system is intended to support veterinarians in assessing the health status of farm animals. Overall, this work demonstrates the potential of using IoT and machine learning for the early detection and diagnosis of skin diseases in farm animals and highlights the importance of continuous monitoring and proactive management in maintaining the health and welfare of these animals.