Object detection system with improved object detection accuracy
First Claim
1. A system for detecting a target object, the system comprising:
- an image pick-up unit configured to pick up an image of a predetermined target region;
a storage unit configured to store therein pattern image data for the target object, the pattern image data representing a feature of the target object;
a similarity determining unit configured to;
(a) set a first block in the picked-up image;
(b) compare a first part of the picked-up image contained in the first block with the pattern image data and repeat the comparison for each of plural shifted locations of the first block in the picked-up image to thereby determine a similarity of respectively corresponding plural parts of the picked-up image contained in a corresponding one of the differently-located first blocks with respect to the pattern image data;
(c) extract, as at least one candidate area, at least one of the differently-located first blocks, if the determined similarity in that block is equal to or greater than a predetermined first threshold similarity;
(d) set a second block in the picked-up image; and
(e) compare a second part of the picked-up image contained in the second block with the pattern image data and repeat the comparison for each of plural shifted locations of the second block in and around the at least one candidate area in the picked-up image to thereby determine a similarity of each of the second parts of the picked-up image contained in a corresponding one of the differently-located second blocks with respect to the pattern image data; and
a specifying unit configured to;
(a) extract some of the differently-located second blocks from all of the differently-located second blocks, if the determined similarity of the second part of the picked-up image contained in each of the extracted differently-located second blocks is equal to or greater than a predetermined second threshold similarity;
(b) calculate, as information indicative of the frequency distribution of some of the differently-located second blocks in the picked-up image, a rate of the number of some of the differently-located second blocks to the number of all of the differently-located second blocks;
(c) determine whether the calculated rate of the number of some of the differently-located second blocks to the number of all of the differently-located second blocks is equal to or greater than a predetermined threshold rate; and
(d) specify the target area in the picked-up image when it is determined that the calculated rate of the number of the extracted differently-located second blocks to the number of all of the differently-located second blocks is equal to or greater than the predetermined threshold rate.
1 Assignment
0 Petitions
Accused Products
Abstract
In a system for detecting a target object, a similarity determining unit sets a block in a picked-up image, and compares a part of the picked-up image contained in the block with a pattern image data while changes a location of the block in the picked-up image to determine a similarity of each part of the picked-up image contained in a corresponding one of the different-located blocks with respect to the pattern image data. A specifying unit extracts some different-located blocks from all of the different-located blocks. The determined similarity of the part of the picked-up image contained in each of some different-located blocks is equal to or greater than a predetermined threshold similarity. The specifying unit specifies, in the picked-up image, a target area based on a frequency distribution of some different-located blocks therein.
24 Citations
4 Claims
-
1. A system for detecting a target object, the system comprising:
-
an image pick-up unit configured to pick up an image of a predetermined target region; a storage unit configured to store therein pattern image data for the target object, the pattern image data representing a feature of the target object; a similarity determining unit configured to; (a) set a first block in the picked-up image; (b) compare a first part of the picked-up image contained in the first block with the pattern image data and repeat the comparison for each of plural shifted locations of the first block in the picked-up image to thereby determine a similarity of respectively corresponding plural parts of the picked-up image contained in a corresponding one of the differently-located first blocks with respect to the pattern image data; (c) extract, as at least one candidate area, at least one of the differently-located first blocks, if the determined similarity in that block is equal to or greater than a predetermined first threshold similarity; (d) set a second block in the picked-up image; and (e) compare a second part of the picked-up image contained in the second block with the pattern image data and repeat the comparison for each of plural shifted locations of the second block in and around the at least one candidate area in the picked-up image to thereby determine a similarity of each of the second parts of the picked-up image contained in a corresponding one of the differently-located second blocks with respect to the pattern image data; and a specifying unit configured to; (a) extract some of the differently-located second blocks from all of the differently-located second blocks, if the determined similarity of the second part of the picked-up image contained in each of the extracted differently-located second blocks is equal to or greater than a predetermined second threshold similarity; (b) calculate, as information indicative of the frequency distribution of some of the differently-located second blocks in the picked-up image, a rate of the number of some of the differently-located second blocks to the number of all of the differently-located second blocks; (c) determine whether the calculated rate of the number of some of the differently-located second blocks to the number of all of the differently-located second blocks is equal to or greater than a predetermined threshold rate; and (d) specify the target area in the picked-up image when it is determined that the calculated rate of the number of the extracted differently-located second blocks to the number of all of the differently-located second blocks is equal to or greater than the predetermined threshold rate. - View Dependent Claims (2, 3, 4)
-
Specification