A method and device for detecting objects in an image
A technology for detecting images and objects, applied in the field of image processing, can solve problems such as slow processing speed and unavailability
Active Publication Date: 2015-09-02
BEIJING VIMICRO ARTIFICIAL INTELLIGENCE CHIP TECH CO LTD
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the existing object detection technology often needs to traverse and search images of multiple scales in order to detect targets of various scales, so the processing speed is very slow and cannot be used in many applications.
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
[0036] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, specific embodiments will be described in detail below with reference to the accompanying drawings.
[0037] Proven object detection methods based on pyramid image scaling and window traversal. The steps of the method are described as follows:
[0038] First, the processed image is scaled to obtain a series of images of different scales. The scale difference between adjacent layers of images is fixed multiples, and the back layer is obtained from the front layer through image scaling. This multiple is called the zoom factor; for example, the first layer is set to be an image with exactly the same size and pixel brightness as the processed image, and the second layer image is the scale of the first layer image scaled to the scale of the first layer image The scale of the image can be 0.8, and the third layer image is an image whose scale is twice the ...
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 provides a method and a device for detecting objects in an image. The method comprises the following steps of: firstly, calibrating a camera to obtain a corresponding relation between the shoot pixel height of an object target in a shot image of the camera and the actual height of the object target; secondly, setting a mask image, and setting the mask pixel height of the object target in the mask image according to the actual height; thirdly, acquiring a classifier model of the object target according to the mask image; and fourthly, adopting a zooming mode to obtain zooming images with different sizes of the shot image, determining a search region of the classifier model in the zooming images according to the mask image height, the actual height and the corresponding relation, and performing searching in the search region by using the classifier model. The method and the device reduce redundant windows to be searched by limiting target searching sizes, improve the searching speed, and solve the technical problem that the prior art has a very slow processing speed.
Description
technical field [0001] The invention relates to image processing technology, in particular to a method and device for detecting objects in images based on camera calibration technology. Background technique [0002] Object detection is the basis of video analysis and has important application value. Compared with object tracking and foreground extraction matching technology, object detection technology has more prominent robustness (robustness, robustness) to illumination and noise interference. However, the existing object detection technology often needs to traverse and search images of multiple scales in order to detect targets of various scales, so the processing speed is very slow and cannot be used in many applications. Contents of the invention [0003] The object of the present invention is to provide a method and device for detecting objects in an image, which reduces redundant windows to be searched by limiting the target search scale, improves search speed, and...
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): G06T7/00
Inventor 邓亚峰谢东海黄英王磊
Owner BEIJING VIMICRO ARTIFICIAL INTELLIGENCE CHIP TECH CO LTD