RTI-Grain: A Method for Detecting the Foreign Body of Granary Based on RSS

This paper proposes a non-destructive, fast, low-cost, non-contact method for detecting stored grain foreign body based on received signal strength (RSS) information , RTI- Grain. The feasibility of stored grain foreign body detection was verified by using RSS data. On this basis, the RTI-Grain detection method was designed. Firstly, the RSSI data is subjected to data normalization, noise elimination and averaging preprocessing, and then ellipse weight model was used to compute the attenuation coefficients in each pixel, finally the Tikhonov regularization was adopted to reconstruct the image of the abnormal area. The experimental verification shows that the foreign body of empty hole, metal and water in granary can be detected, and the detection accuracy is 98.1% at maximum.