Computationally Efficient Implementation of Convolution-based Locally Adaptive Binarization Techniques
This work addresses the problem of enabling efficient binarization on devices with limited computing facilities, representing an incremental improvement over existing methods.
The paper tackled the high computational cost of locally adaptive binarization techniques for document image processing by presenting a computationally efficient implementation that reduces complexity from O(W^2N^2) to O(WN^2), achieving a 5 to 15 times reduction in computation time while maintaining comparable performance.
One of the most important steps of document image processing is binarization. The computational requirements of locally adaptive binarization techniques make them unsuitable for devices with limited computing facilities. In this paper, we have presented a computationally efficient implementation of convolution based locally adaptive binarization techniques keeping the performance comparable to the original implementation. The computational complexity has been reduced from O(W2N2) to O(WN2) where WxW is the window size and NxN is the image size. Experiments over benchmark datasets show that the computation time has been reduced by 5 to 15 times depending on the window size while memory consumption remains the same with respect to the state-of-the-art algorithmic implementation.