• Ensuring Automated Security in Close Water (Pond) Fish Cultivation.

Details: In our study, we proposed a novel technique to detect and supervise illegal fishing by using automatic video surveillance technique. This technique aims at detecting the presence of abandoned objects in a protected fisheries area and at automatically performing online semantic video segmentation in order to facilitate the human operator’s task of retrieving the cause of an alarm. This task is performed by operating image segmentation based on temporal rank-order filtering followed by classification using convolutional neural network (CNN) in order to reduce false alarms. In our system, the key frame is the one whether an object like human with or without fishhook or net is moving in protected fisheries area. We used image processing technique for detecting the foreground object and deep learning for classification of moving object.

• Automatic paddy seed variety identification using Haralick texture features.

Details: This research is aimed at developing an automated system to identify paddy varieties by using multiple heterogeneous features extracted exploiting external, physical and textural properties. The external and physical features are extracted using the color and physical properties of paddy seed while the textural features are estimated by Haralick technique. These features are utilized to train feed forward neural network (FNN) to predict the paddy variety.

• IoT Enabled Agricultural Storage Monitoring.

Details: In our study, we proposed a system to monitor all agricultural item during the period of storage. It informs the user if the storage suffers from any difficulties like abrupt temperature or humidity changes.