When infrared focal plane array imaging system detects targets, especially small targets, there is the problem of low gray resolution. In this paper, an adaptive scene-based gray super-resolution technique is proposed, aiming to solve the problem. The paper gives a detailed description on the method of image gray super-resolution by adjusting the signal sample range in infrared focal plane array (IRFPA) imaging system. The method contains the following three parts: extracting the effective gray range from the scene, and obtaining the basis of super-resolution adjustment; providing the adjusting parameters after filter-predicting the basis of adjustment, combining with the adaptive LMS-based filtering algorithm; and completing gray super-resolution by controlling the parameters in super-resolution circuit. Finally, the total solution is experiment validated. The experiment in infrared focal plane array imaging system has proven the feasibility and effectiveness of this method, and the improvement of super-resolution. Then test set shows the MRTD can be increased more than one time.
Interpolation is a necessary processing step in 3-D reconstruction because of the non-uniform resolution. Conventional interpolation methods simply use two slices to obtain the missing slices between the two slices .when the key slice is missing, those methods may fail to recover it only employing the local information .And the surface of 3D object especially for the medical tissues may be highly complicated, so a single interpolation can hardly get high-quality 3D image. We propose a novel binary 3D image interpolation algorithm. The proposed algorithm takes advantages of the global information. It chooses the best curve adaptively from lots of curves based on the complexity of the surface of 3D object. The results of this algorithm are compared with other interpolation methods on artificial objects and real breast cancer tumor to demonstrate the excellent performance.