| In this paper the Haar wavelet is used for decomposition and reconstruction of an image. Haar wavelet transform is based on an averaging and differencing algorithm. To calculate the Haar wavelet transform of an image matrix, we consider each row of a matrix. In this row we find the average of each pair of samples, and then find the difference between the average and the samples. After finding the averages and differences, fill the first half of the array with averages and second half with differences and again repeat the process on the first half of the array. At the end of this process we get an image matrix in the new domain. If we take the inverse Haar wavelet transform of this image matrix, we can easily reconstruct the original image.; This is implemented in real-time on TI's TMS320C6701 using C programming and assembly language programming. Finally, we analyze the code by profiling it in Code Composer Studio and counting the CPU cycles. |
