How is the algorithm bogus (assuming that there isn't a mistake somewhere and that the results are validated)? Just because the output in the examples is similar to other edge detectors doesn't mean the work is 'bogus' or worthless. It's a new algorithm, a new way of doing things. It may have advantages in specific areas over your 'trivial contrast-enhancing filters'. For example, there are many sorting algorithms, but you can get the same results regardless of the algorithm. Does this mean that any sorting algorithm other than the most trivial is 'bogus'?