Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A positioning method, image moment technology, applied in the field of image processing
Active Publication Date: 2014-12-03
BEIJING INST OF CONTROL ENG
View PDF2 Cites 2 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0007] The technical problem solved by the present invention is: to overcome the shortcomings of existing target positioning methods, and provide a high-prec...
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 more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0052] (1) Take a one-dimensional Gaussian signal For example, let I 0 =1,σ=1,ε=2e -2 , the pixel scale is 1, the abscissa axis is [-0.5,0.5], take any point X initial Generate a discrete one-dimensional histogram P(i) for the center.
[0053] (2) Use the first-order moment method to obtain the centroid of the one-dimensional histogram P(i) Order X 0 =X m , with X 0 is the initial center point.
[0054] (3) According to the principle of image similarity, the Gaussian weighting function is selected, and its parameter setting is the same as that of the Gaussian signal function in step (1), and the weighting coefficient is calculated
[0055] (4) Use the formula Calculate X 1 , compare|X 1 -X 0 |0 =X 1 , return to step (2) to calculate the weighting coefficient, and so on until |X 1 -X 0 |1 for X final .
[0056] (5) Change X initial value, repeat steps (1)-(4). Define the curve "first moment method" as X m -X initial , represents the centroid extraction ...
Embodiment 2
[0061] The method is the same as the step of embodiment 1, except that the initial centroid position adopted by the method of the present invention in step (1) is the center of the brightest pixel, and all the other parameters are the same as above, and the obtained results are shown in Figure 4 . From Figure 4 It can be seen that the centroid error of all sample points obtained by the method of the present invention is better than that of the first-order moment method.
Embodiment 3
[0063] The method is the same as the step of embodiment 1, the difference is that the power weighting function ω(x)=1 / (1+|x-X 0 |), add 10% intensity white noise, the other parameters are the same as above, the obtained results are shown in Figure 5 . From Figure 5 It can be seen that, except for a few sample points, the centroid error obtained by the method of the present invention is better than that of the first-order moment method for most of the sample points.
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 a high-precision image moment positioning method, which comprises the following steps: (1) separating an image of a target to be detected from background; (2) calculating the initial moment of the target; (3) selecting a weighting function suitable for target image distribution as a weighting coefficient, wherein the weighting function needs to exhibit four characteristics of normalization, centrosymmetry, attenuating property and incoherence; (4) according to the weighting coefficient, recalculating a new moment of the target image, outputting the new moment as a target centroid if an error between the new moment and an original moment satisfies precision requirements; otherwise, taking the new moment as an iteration starting point to reselect the weighting coefficient and recalculate the new moment until the error between the new moment and the original moment satisfies the precise requirements; and outputting the new moment as a target parameter. The method has the advantages of being high in target positioning precision, high in anti-noise capability, simple in calculation and easy for engineering realization, does not aim at a specific target image and is especially suitable for occasions which adopt the image with a low signal to noise ratio and complex background to carry out target positioning.
Description
technical field [0001] The invention relates to an image processing method, which is especially suitable for the application occasions of target positioning using low signal-to-noise ratio and complex background images. Background technique [0002] Object tracking is a typical problem in the field of computer vision, and one of its core technologies is how to perform high-precision image positioning on the tracked object. [0003] At present, the following methods are mainly used for image pose positioning: that is, the pixel gray distribution of the target image is directly used. Such as the direct moment method and the direct moment method with threshold, the advantage is that the gray information of each pixel is fully utilized, and the calculation is simple, and the direct moment method with threshold uses image preprocessing to remove part of the background noise, and the accuracy is higher than that of the direct moment method. will improve. The disadvantage is that...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.