Depth Detection Method and System Using Thereof

• US 20110141274A1
• Filed: 07/23/2010
• Published: 06/16/2011
• Est. Priority Date: 12/15/2009
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Liao, Chih-Pin, Wang, Ko-Shyang, Tsai, Yao-Yang, Huang, Jay
• Granted Patent: US 8,525,879 B2
• Time in Patent Office: Days
• US Class Current: 348/139
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A depth detection method includes the following steps. First, first and second video data are shot. Next, the first and second video data are compared to obtain initial similarity data including r×c×d initial similarity elements, wherein r, c and d are natural numbers greater than 1. Then, an accumulation operation is performed, with each similarity element serving as a center, according to a reference mask to obtain an iteration parameter. Next, n times of iteration update operations are performed on the initial similarity data according to the iteration parameter to generate updated similarity data. Then, it is judged whether the updated similarity data satisfy a character verification condition. If yes, the updated similarity data is converted into depth distribution data.

Claims

1. A depth detection system, comprising:

a dual camera apparatus for shooting first video data and second video data, which respectively correspond to a first viewing angle and a second viewing angle, wherein each of the first and second video data comprise r×

c sets of pixel data, wherein r and c are natural numbers greater than 1;

a horizontal calibration apparatus for performing horizontal calibration on the first and second video data, and outputting the first and second video data, which are horizontally calibrated; and

a depth estimation apparatus, comprising;

a similarity estimation module for comparing pixel data of the first and second video data, provided by the horizontal calibration apparatus, with each other to obtain initial similarity data, which comprise r×

c sets of initial similarity elements, each comprising d initial similarity elements, wherein d is a natural number greater than 1;

an iteration update module for selecting multiple initial similarity elements to perform an accumulation operation to obtain an iteration parameter according to a reference mask with each of the initial similarity elements serving as a center, wherein the iteration update module performs n times of iteration update operations on the initial similarity data according to the iteration parameter to generate updated similarity data, which comprise r×

c sets of update similarity elements, each comprising d similarity elements; and

a control module for judging whether each of the r×

c sets of update similarity elements satisfies a character verification condition;

wherein when the r×

c sets of update similarity elements satisfy the character verification condition, the control module converts the r×

c sets of update similarity elements into depth distribution data.

2. The system according to claim 1, wherein when the r×

c sets of update similarity elements do not satisfy the character verification condition, the control module adjusts a size of the reference mask, and drives the iteration update module to again calculate the iteration parameter according to the adjusted reference mask and again generate the updated similarity data comprising the r×

c sets of update similarity elements according to the recalculated iteration parameter.

3. The system according to claim 2, wherein:

the control module further judges whether the re-obtained r×

c sets of update similarity elements satisfy the character verification condition; and

when each of the re-obtained r×

c sets of update similarity elements satisfies the character verification condition, the control module converts the re-obtained r×

c sets of update similarity elements into the depth distribution data.

4. The system according to claim 3, wherein when each of the re-obtained r×

c sets of update similarity elements still does not satisfy the character verification condition, the control module again adjusts the size of the reference mask, and drives the iteration update module to again calculate the iteration parameter according to the adjusted reference mask, and again generate the updated similarity data comprising the r×

c sets of update similarity elements according to the recalculated iteration parameter.

5. The system according to claim 1, further comprising:

a characteristic analyzing apparatus for receiving and analyzing the first and second video data provided by the horizontal calibration apparatus to obtain first characteristic region data and second characteristic region data according to the first and second video data, respectively, wherein the first characteristic region data respectively correspond to the second characteristic region data; and

a depth range estimation apparatus, which comprises;

an estimation module for receiving and calculating a horizontal displacement quantity between the first characteristic region data and the corresponding second characteristic region data, and converting the horizontal displacement quantity into depth data;

a statistics module for converting the depth data into a set of depth statistics distribution data; and

an operation module for obtaining a minimum depth value and a maximum depth value from the set of depth statistics distribution data according to a first critical condition, and determining depth range data corresponding to the first and second video data according to the minimum and maximum depth values;

wherein the depth estimation apparatus determines the depth distribution data based on the depth range data.

6. The system according to claim 5, wherein the operation module obtains the minimum and maximum depth values from the depth statistics distribution data according to a critical number.

7. A depth detection method, comprising the steps of:

shooting first video data and second video data, which respectively correspond to a first viewing angle and a second viewing angle, wherein each of the first and second video data comprise r×

c sets of pixel data, wherein r and c are natural numbers greater than 1;

performing horizontal calibration on the first and second video data;

comparing pixel data of the horizontally calibrated first and second video data with each other to obtain initial similarity data, wherein the initial similarity data comprise r×

c sets of initial similarity elements, and each of the r×

c sets of initial similarity data comprise d initial similarity elements, wherein d is a natural number greater than 1;

selecting multiple similarity elements according to a reference mask with each of the similarity elements serving as a center, and performing an accumulation operation on the selected similarity elements to obtain an iteration parameter;

performing n times of iteration update operations on the initial similarity data according to the iteration parameter to generate r×

c sets of update similarity elements, each comprising d similarity elements;

judging whether each of the r×

c sets of update similarity elements satisfies a character verification condition; and

converting the r×

c sets of update similarity elements into depth distribution data when the r×

c sets of update similarity elements satisfy the character verification condition.

8. The method according to claim 7, further comprising the step of:

adjusting a size of the reference mask when the r×

c sets of update similarity elements do not satisfy the character verification condition;

wherein after the size of the reference mask is adjusted, the steps of obtaining the iteration parameter, generating the r×

c sets of update similarity elements and judging whether each of the r×

c sets of update similarity elements satisfies the character verification condition are repeated.

9. The method according to claim 8, further comprising the step of:

converting the re-obtained r×

c sets of update similarity elements into the depth distribution data when each of the re-obtained r×

c sets of update similarity elements satisfies the character verification condition.

10. The method according to claim 9, further comprising the step of:

repeating the step of adjusting the size of the reference mask when each of the re-obtained r×

c sets of update similarity elements still does not satisfy the character verification condition;

wherein after the size of the reference mask is adjusted, the steps of obtaining the iteration parameter, generating the r×

c sets of update similarity elements and judging whether each of the r×

c sets of update similarity elements satisfies the character verification condition are repeated.

11. The method according to claim 7, further comprising, before the step of obtaining the initial similarity data, the steps of:

analyzing the horizontally calibrated first and second video data to obtain first characteristic region data from the first video data and to obtain second characteristic region data from the second video data, wherein the first characteristic region data correspond to the second characteristic region data;

calculating a horizontal displacement quantity between the first characteristic region data and the corresponding second characteristic region data, and converting the horizontal displacement quantity into depth data;

converting the depth data into one set of depth statistics distribution data; and

obtaining a minimum depth value and a maximum depth value from the set of depth statistics distribution data according to a first critical condition, and determining depth range data corresponding to the first and second video data according to the minimum and maximum depth values.

12. The method according to claim 11, wherein an operation module obtains the minimum and maximum depth values from the depth statistics distribution data according to a critical number.