Vlsi for Moment Computation and Its Application to Breast Cancer Detection

Fast and scalable computations of 2D image moments

“…In addition, three cost optimal algorithms are also derived on the AROB model, the hypercube computer and the EREW PRAM model in Oð N 2 t 2 þ N log t t Þ time using t 2 processors, respectively, where 1 6 t 6 N. To the best of our knowledge, there were no algorithms before which could solve this problem in constant time and/or provide such a performance on any architectures. Clearly, our results achieve time or cost optimal and are better than the previously known algorithms proposed in the literature [2,3,20]. In addition, we also develop a parallel implementation on an IBM SP2 to verify the performance of the proposed parallel algorithms.…”

“…The proposed algorithms have the following advantages. O(log N) Cheng [3] Array N + 1 O(N) N · (N + 1) O(N) This paper…”

“…At each step, every processor performs the same instruc- Table 2 The coefficients for the first 10 moments p a (p,1) a (p,2) a (p, 3) a (p,4) a (p,5) a (p,6) a (p,7) a (p, 8) a (p, 9) a (p,10) 1 1 2 1 1 3 1 4 1 4 1 11 11 1 5 1 26 66 26 1 6 1 57 302 302 57 1 7 1 120 1191 2416 1191 120 1 8 1 247 4293 15619 15619 4293 247 1 9 1 502 14608 88234 156190 88234 14608 502 1 10 1 1013 47840 455192 1310354 1310354 455192 47840 1013 1 tion, with a number of processors marked out. There are three kinds of shared memory PRAM models.…”

“…Chen's algorithms run in O(N) time on a linear array of size N and run in O(log N) time on a 2D array of size N · N, respectively. Cheng et al [3] proposed two special VLSI architectures for computing the 2D moments of order (p + q). For the 1D structure, it took O(max(p, q) · N) time using N + 1 processors.…”

Image indexing by Color Plane Moment

“…Since introduced by Hu in 1961, the method of moment invariant has become one of the most important techniques for image analysis. Its successful applications have been reported in areas such as aircraft identification, remotely sensed data matching, character recognition, edge detection, pavement distress classification, breast cancer detection, and so on (Cash and Hatamian, 1987;Cheng et al, 1998;H. Chou and Chen, 1990;J.…”