초록 ▼

A Fine-Grained Microstructure Processor (FMP) (10) for solving assignment and correlation problems that utilizes a pair of arrays (22, 24) of cells (26) to represent the position of two set of two-dimensional data points (14). The arrays (22, 24) are divided by a predefined shaped (16) having a plur

A Fine-Grained Microstructure Processor (FMP) (10) for solving assignment and correlation problems that utilizes a pair of arrays (22, 24) of cells (26) to represent the position of two set of two-dimensional data points (14). The arrays (22, 24) are divided by a predefined shaped (16) having a plurality of regions (18). The FMP (10) counts the number of data points (14) in each region (18) and then finds the difference between the resulting sums from corresponding regions (18) in the two arrays (22, 24). The differences for each regions (18) are added together by an accumulator (34) to determine a correlation factor. The data in the second array (24) is then shifted until a different data point (14) occupies a particular position with respect to the predefined shape (16) and a new correlation factor is found. The FMP (10) finds a correlation factor for each data point (14) in the second array (24) and the lowest correlation factor represents the best match between a data point (14) in the first array (22) at a particular position with respect to the predefined shape (16), and the point in the second array (24) yielding the best match. The FMP (10) may then shift the data in the first array (22) and repeat the above process to match each data point (14) in the first array (22) with a data point (14) in the second array (24).

대표청구항 ▼

A method for associating a first set of data points with a second set of data points, said method comprising: a. locating each data point in the first set at a given position in a first two-dimensional array, and locating each data point in the second set at a given position in a second two-dimensio

A method for associating a first set of data points with a second set of data points, said method comprising: a. locating each data point in the first set at a given position in a first two-dimensional array, and locating each data point in the second set at a given position in a second two-dimensional array; b. defining a plurality of two-dimensional regions having predetermined shapes within the first and second arrays, said regions in the first array having corresponding regions with the same shape in the second array; c. defining a single position within each of said two-dimensional arrays having a fixed relationship with said regions, said data points being located in said first and second array so that one data point is at the single position; d. adding the number of data points in each region; e. for each of said regions in said first array, subtracting the total number of data points in said region, from the total number of data points in the corresponding region of the second array; f. adding up all said differences for each region to arrive at a correlation factor; g. moving said data points in the second array, keeping the data points fixed with respect to each other until a different data pint occupies the single position, and repeating steps d-f until a new correlation factor is derived; h. repeating step g for every point int he second array so that a correlation factor is derived for each data pint in the second array when it occupies the single position; i. finding the minimum of said correlation factors; and j. identifying the data point that occupies the single position when the minimum correlation factor was derived, wherein the point at the single position in the first array is associated with the data point in the single position int he second array when the minimum correlation factor was derived.