An interpolation method for missing data in depth images
A technology for missing data and depth images, which is applied in the field of image processing, can solve problems such as poor repair of image processing methods, missing data, missing data, etc., achieve strong real-time operation efficiency, high estimation accuracy, and meet the requirements of valuation and The effect of imputation requirements
Active Publication Date: 2017-12-22
深圳杉谷科技有限公司
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In order to solve the technical problem that a large amount of data is missing in the depth image collected by the depth image sensor in the prior art, and the common image processing method cannot be well repaired, the present invention is an interpolation method for the missing data of the depth image, which includes the following steps:
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0033] First of all, it needs to be explained that the depth image processed in this specific embodiment comes from the Kinect depth image sensor, and the data includes depth image and color image:
[0034] (1) Depth image: The real-time depth image within the shooting range at time t is represented by D(x, y, t), the image size is 640*480, and the range of the pixel depth value (that is, the pixel value in the depth image) is [0 -8096], indicating the distance from the point in the space to the camera plane, where the pixel with a depth value of 0 represents the pixel with missing data, that is, the pixel whose depth value of the pixel needs to be re-estimated and interpolated by the present invention.
[0035] (2) Color image: The real-time color image within the shooting range corresponding to the depth image at time t is represented by I(x, y, t), the image size is 640*480, and the color image includes three channels of R, G, and B. The range of pixel values in each chan...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an interpolation method for missing data in a depth image, which comprises the following steps: 1. Extracting all depth sub-images containing missing data pixels for pixels with missing data in the depth image; According to the ratio of the pixels of the data to the total pixels, all depth sub-images are divided into two categories; 3. For the two types of depth sub-images, the estimation method based on temporal and spatial neighborhood information clustering analysis and the local neighborhood maximum The method of estimating the value is to estimate the missing data; 4. For the depth image, replace the missing data with the estimated value obtained according to the third step to obtain the repaired depth image. The invention not only has high estimation accuracy, but also has strong real-time operation efficiency, and can meet the estimation and interpolation requirements of large-area missing data in depth images.
Description
technical field [0001] The invention relates to an image processing method, in particular to a method for estimating and interpolating missing data in a depth image acquisition process. Background technique [0002] With the advancement of science and technology, depth image sensors have been widely used in human-computer interaction, robot navigation, object recognition, 3D modeling and other fields. The depth image sensor mainly adopts the real-time acquisition technology of stereo vision image based on "light-coding", which can acquire the RGB-D image in the target area in real time (that is, the image containing both color RGB information and depth information). Its main product is Microsoft The company's Kinect is the representative, as well as ASUS's Xtion-Pro and other similar products. [0003] "Light-coding" is a depth measurement technology developed by PrimeSense. The depth image sensor based on "light-coding" technology mainly includes an infrared laser speckle...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/00G06T3/40
Inventor 王江斌程元火夏鹏缪金梁
Owner 深圳杉谷科技有限公司