Method of detecting focus position of object from variable gray level image of object

Claim 2

A focus position detecting method comprising steps of: obtaining quantized two-dimensional image data including an object to be in focus with respect to an optical system by imaging the object through the optical system; defining a specified area in the two-dimensional image data; defining a partial image in the specified area, the partial image being composed of a target pixel and peripheral pixels on a line in a specified direction in which the target pixel exists; defining a left area and a right area respectively positioned on left and right sides of the target pixel on the line in the specified direction; calculating gray-level data of each pixel in the left area; calculating gray-level data of each pixel in the right area; obtaining as a gray-level ratio a ratio of calculation result of the left area to calculation result of the right area; calculating an evaluation function by means of the gray-level ratio by repeating the gray-level ratio calculating step in the specified area; obtaining a plurality of evaluation function calculation values in a variety of optical positions by repeating the steps for calculating the gray-level data of the left and right areas, obtaining the gray-level ratio, and calculating the evaluation function through variation of the optical position of the object; and obtaining a focus position of the object with respect to the optical system based on the evaluation function calculation values; wherein after an average value AV1 of gray-level data of each pixel in the right area and an average value AV2 of gray-level data of each pixel in the left area are calculated, the gray-level ratio of the average value AV1 to the average value AV2 is calculated by the following condition: when the average value AV1 of the right area is larger than the average value AV2 of the left area, the gray-level ratio is AV1/AV2; when the average value AV1 of the right area is smaller than the average value AV2 of the left area, the gray-level ratio is AV2/AV1; and when the average value AV1 of the right area is the same as the average value AV2 of the left area, the gray-level ratio is 1.0.

Claim 3

The focus position detecting method as claimed in claim 1, wherein a position having the greatest cumulative gray-level ratio is selected as the focus position.

Claim 4

The focus position detecting method as claimed in claim 1, wherein interpolation data are obtained for the plural number of evaluation function calculation values in a plurality of optical positions, and the focus position of the object with respect to the optical system is obtained based on a result including the interpolation data.

Claim 5

A focus position detecting method comprising steps of: obtaining quantized two-dimensional image data including an object to be in focus with respect to an optical system by imaging the object through the optical system; defining a specified area in the two-dimensional image data; defining a partial image in the specified area, the partial image being composed of a target pixel and peripheral pixels on a line in a specified direction in which the target pixel exists; defining a left area and a right area respectively positioned on left and right sides of the target pixel on the line in the specified direction; calculating gray-level data of each pixel in the left area; calculating gray-level data of each pixel in the right area; obtaining as a gray-level ratio a ratio of calculation result of the left area to calculation result of the right area; calculating an evaluation function by means of the gray-level ratio by repeating the gray-level ratio calculating step in the specified area; obtaining a plurality of evaluation function calculation values in a variety of optical positions by repeating the steps for calculating the gray-level data of the left and right areas, obtaining the gray-level ratio, and calculating the evaluation function through variation of the optical position of the object; and obtaining a focus position of the object with respect to the optical system based on the evaluation function calculation values; wherein after an average value AV1 of gray-level data of each pixel in the right area and an average value AV2 of gray-level data of each pixel in the left area are calculated, the gray-level ratio of the average value AV1 to the average value AV2 is calculated by the following condition: when the average value AV1 of the right area is larger than .the average value AV2 of the left area, the gray-level ratio is AV2/AV1; when the average value AV1 of the right area is smaller than the average value AV2 of the left area, the gray-level ratio is AV1/AV2; and when the average value AV1 of the right area is the same as the average value AV2 of the left area, the gray-level ratio is 1.0.

Claim 6

A focus position detecting method comprising steps of: obtaining quantized two-dimensional image data including an object to be in focus with respect to an optical system by imaging the object through the optical system; defining a specified area in the two-dimensional image data; defining a partial image in the specified area, the partial image being composed of a target pixel and peripheral pixels on a line in a specified direction in which the target pixel exists; defining a left area and a right area respectively positioned on left and right sides of the target pixel on the line in the specified direction; calculating gray-level data of each pixel in the left area; calculating gray-level data of each pixel in the right area; obtaining as a gray-level ratio a ratio of calculation result of the left area to calculation result of the right area; calculating an evaluation function by means of the gray-level ratio by repeating the gray-level ratio calculating step in the specified area; obtaining a plurality of evaluation function calculation values in a variety of optical positions by repeating the steps for calculating the gray-level data of the left and right areas, obtaining the gray-level ratio, and calculating the evaluation function through variation of the optical position of the object; and obtaining a focus position of the object with respect to the optical system based on the evaluation function calculation values; wherein after an average value AV1 of gray-level data of each pixel in the right area and an average value AV2 of gray-level data of each pixel in the left area are calculated, the gray-level ratio of the average value AV1 to the average value AV2 is calculated by the following condition: when the average value AV1 of the right area is larger than .the average value AV2 of the left area, the gray-level ratio is AV2/AV1; when the average value AV1 of the right area is smaller than the average value AV2 of the left area, the gray-level ratio is AV1/AV2; and when the average value AV1 of the right area is the same as the average value AV2 of the left area, the gray-level ratio is 1.0.

Claim 7

The focus position detecting method as claimed in claim 4, wherein a position having the smallest cumulative gray-level ratio is selected as the focus position.

Claim 8

The focus position detecting method as claimed in claim 4, wherein interpolation data are obtained for the plural number of evaluation function calculation values in a plurality of optical positions, and the focus position of the object with respect to the optical system is obtained based on a result including the interpolation data.