An image processing resolution enhancement method for single-pixel targets

A resolution enhancement technique for image processing systems is advantageous for applications limited to small sensor arrays or when relatively high resolution is required. The spatial quantization of image sensors causes a nonlinear input-output relationship which is a basic limitation. A new artificial means of perturbing the sensor array is presented that enhances resolution by smoothing the nonlinear input-output characteristic. Simulation results demonstrate that effective quantization accuracy can be improved by over two orders of magnitude in two dimensions. The method is limited to systems in which the target image is approximately single-pixel in size and of uniform intensity. System implementations using the Hough transform meet the requirements for the new method.< >