Abstract:
You can read the entire findings of my proposal, a "Chicken disease detection web app using AI," here. This article goes into detail on the processes used to develop the idea into a working website. The user dashboard represents a particular element that sticks out to system users. Chicken diseases are very harmful for every chicken farm for their business. Because chicken producers have little opportunity for agricultural support services, these illnesses do not show symptoms in their flocks of birds early. Deep learning methods may be able to identify certain poultry illnesses early on. This website will require user input in order to identify chicken diseases. Thus, there will be two basic components to it. One involves developing websites, while the other involves using deep learning to identify chicken diseases. In this work, there have been used 4 classes of chicken diseases for detection using web application like: “Salmonella”, “Coccidiosis”, “New Castle disease” and “Healthy”. To predict and detect chicken sickness in order to categorize utilizing chicken coop diseases, four models are used: CNN, VGG16, MobileNetV3, and Resnet. Resnet achieves the greatest accuracy of 81.70%. Ultimately, the Resnet network is used for classification to identify chicken sickness and provide a web app. The report covers all aspect of the web application's creation procedure, from conception to execution, including its structure, user interface fashion, and the technologies used. We used Django with Python for the backend, JavaScript was used for the user interface. Our system application may be set up with just a standard desktop machine and internet access; costly software or computer components are not required.
