A hybrid mobile environmental and population density management system for smart poultry farms

•We proposed the hybrid model of mobile phone and WSNs for smart poultry farm.•We proposed the use of cloud services as database and computational offloading.•We optimized the transmission logics to monitor/control the environment behavior.•We proposed the methodology for image filer to classify the poultry population.

Due to advances in technology and size reduction of portable mechanical and electronic devices, sensors have been involved in many sectors, including in agriculture, where sensors and their unique functionalities are extremely useful for both productivity gains and reducing operating costs. These benefits are attained by utilizing their real-time sensing capability over environmental stages affecting the animal breed to enable on-time decision support but with some limitations, i.e., mobility, coverage, ubiquitous access, and energy. Thus, this research focuses on the integration of wireless sensor and mobile system networks with a well-known sensor integration platform toward cloud offloading scalability services via a hybrid architecture used to collect sensing data, such as temperature, humidity, light intensity, and population density, for data analytics and then issuing on-time decisions to adjust the environmental behavior accordingly. Based on a smart poultry farm concept for evaporative cooling environments, the instrument and components of the system design are discussed in detail with the experienced selection criteria, including a discussion of practical topology and deployments, enhanced transmission logics, external environmental tuning control logics integrating mobile user management interfaces, and image processing units. Aside from the proposed prototype of the integration of mobile phones, sensors, and controllers, an experimental investigation was also performed on data sensing and transmission procedures regarding power consumption characteristics, especially on high-cost image data transmissions, including an illustration of the outstanding performance, i.e., 80% in accuracy with low computational complexity, of the image filter over other well-known image classification techniques.