Image Restoration Of Reflective Area In Organs Of Diseased Pigs Based On Improved Criminisi Algorithm

The moisture content present in the organs of diseased pigs leads to the reflective phenomenon when collecting images. The resulted problem thus affects the recognition of the symptoms of the diseases. In order to improve the reliability of identification, an improved algorithm was proposed in this research based on the Criminisi algorithm. The artificial markup was processed with mathematical morphology closing operation, so that the area to be repaired could be completely covered by the marker. This ensures the correctness of the markup and improves the repair result. The normalizing function is introduced to define and to avoid the fast decay of the confidence term. The priority function is defined in the form of weighted sum, which avoids the problem of unreliable repair order caused by zero priority value. The experimental results show that the improved Criminisi algorithm is more reliable and the filling results look more natural and realistic.