IP Library Granted Patent US 10,257,405
Granted Patent B2
US 10,257,405 · App. 14/986,850 · Granted Apr 9, 2019

Automatic focusing method and automatic focusing system

Inventors: Pi-Lun Cheng (New Taipei, TW); Hsuan-Kuei Huang (New Taipei, TW)
Assignee: NANNING FUGUI PRECISION INDUSTRIAL CO., LTD.
H04N5/23212G06K9/00369G06K9/4652G06K9/4661G06K9/52G06K2009/4666G06T2207/30196
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,257,405
App. No.
14/986,850
Granted
Apr 9, 2019
Kind
B2
Abstract

An automatic focus method includes: catching an image data in a focus area; calculating the image data to get a contrast value curve via a contrast value algorithm; obtaining a contrast weight of each contrast value; calculating an adapted contrast value by multiplying the contrast value to the contrast weight; and adapting the focus according to the adapted contrast value curve. An automatic focus system using the automatic focus method is further disclosed.

Claims (48)

1. An automatic focusing method comprising:

acquiring an image data in a focus area;

calculating the image data to get a contrast value curve by a contrast value algorithm;

acquiring a contrast weight of each contrast value, wherein acquiring a contrast weight of each contrast value comprising:

determining whether the focus area is a skin area;

decreasing contrast weight corresponding to the non-skin area;

determining whether the skin area is integrated;

increasing contrast weight corresponding to the integrated skin area;

determining whether the non-integrated skin area accords with human feature;

increasing the corresponding contrast weight when the non-integrated skin area accords with human feature;

decreasing the corresponding contrast weight when the non-integrated skin area does not accord with human feature;

calculating an adapted contrast value by multiplying the contrast value to the contrast weight; and

adapting the focus according to the adapted contrast value curve.

2. The automatic focusing method of claim 1 , wherein determining whether the focus area is a skin area comprising:

setting a determining area;

converting each pixel in the focus area to the HIS model, getting a hue angle of each pixel;

comparing each hue angle located in the determining area;

confirming the focus area a skin area when the hue angle is located in the determining area; and

confirming the focus area a non-skin area when the hue angle is not located in the determining area.

3. The automatic focusing method of claim 1 , wherein determining whether the skin area is integrated comprising:

building a scan area in a center coordinates of a center of the skin area;

determining whether there is one or two arm areas in the scan area;

confirming the scan area is a skin area when there is one or two arm areas in the scan area; and

confirming the scan area is a non-skin area when there is no arm area in the scan area.

4. The automatic focusing method of claim 3 , wherein building a scan area in a center coordinates of a center of the skin area further comprising: finding out the center of the skin area to be the center coordinates (xc, yc); extending a preset length downwards; respectively extending left and right a preset shoulder length to get two preset shoulder coordinates (xl, yl) and (xr, yr); respectively drawing a circle in a center of circle of (xl, yl) and (xr, yr) and a radius of a preset arm length; and confirming the set of the two circles to the scan area.

5. The automatic focusing method of claim 4 , wherein determining whether the skin area is integrated further comprising: adjusting the scan area by adjusting the preset length, the preset shoulder length, or/and the preset arm length when there is more than two arm areas.

6. The automatic focusing method of claim 1 , wherein determining whether the non-integrated skin area accords with human feature comprising:

determining whether the skin area is shining;

decreasing the corresponding contrast weights when the skin area is shining;

determining whether the non-shining area is moving;

increasing the corresponding contrast weights when the non-shining area is moving; and

decreasing the corresponding contrast weights when the non-shining area is not moving.

7. An automatic focusing system comprising:

an image acquiring module, configured to acquire an image data in a focus area; and

a processor comprising:

a calculating module, configured to calculate the image data to get a contrast value curve by a contrast value algorithm;

a weight acquiring module, configured to acquire a contrast weight of each contrast value, wherein the weight acquiring module comprising:

a skin area confirming module, configured to determine whether the focus area is a skin area;

a weight adjusting module, configured to decrease contrast weights corresponding to the non-skin area;

a skin integrity confirming module, configured to determine whether the skin area is integrated;

a human feature confirming module, configured to determining whether the non-integrated skin area accords with human feature, wherein the weight adjusting module is further configured to increase contrast weights corresponding to the integrated skin area; increase the corresponding contrast weights when the non-integrated skin area accords with human feature; and decrease the corresponding contrast weights when the non-integrated skin area does not accord with human feature;

a revising module, configured to calculate an adapted contrast value by multiplying the contrast value to the contrast weight; and

an adjusting module, configured to adapt the focus according to the adapted contrast value curve.

8. The automatic focusing system of claim 7 , wherein the skin area confirming module comprising:

a setting module, configured to set a determining area;

a converting module, configured to convert each pixel in the focus area to the HIS model, getting a hue angle of each pixel;

a comparison module, configured to compare whether each hue angle is located in the determining area; and

an angle determining module, configured to confirm the focus area a skin area when the hue angle is located in the determining area, and confirm the focus area a non-skin area when the hue angle is not located in the determining area.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 27, 2018
From: NANNING FUGUI PRECISION INDUSTRIAL CO., LTD.; HON HAI PRECISION INDUSTRY CO., LTD.
To: NANNING FUGUI PRECISION INDUSTRIAL CO., LTD.
Reel/Frame 045171/0347 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 4, 2016
From: CHENG, PI-LUN; HUANG, HSUAN-KUEI
To: NANNING FUGUI PRECISION INDUSTRIAL CO., LTD.; HON HAI PRECISION INDUSTRY CO., LTD.
Reel/Frame 037399/0168 →
Priority Claims (1)
TW 104141832 A · Dec 11, 2015 · national
Continuity (1)
Related Publication 20170171457A1 · Jun 15, 2017