dcm-htm haze removal method for optical remote sensing images
An optical remote sensing and imaging technology, applied in the field of haze removal of optical remote sensing images, which can solve the problems of effective image processing of difficult features, poor algorithm versatility, and estimation.
Active Publication Date: 2018-06-22
KUNSHAN DIGITAL CITY INFORMATION TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The main disadvantage of the algorithm is that it is difficult to estimate HTM by obtaining dark targets when dealing with high-reflection background images such as deserts and snow fields. In addition, the method requires band synthesis and is not suitable for images lacking short-band
[0006] As mentioned above, the existing dehazing image processing algorithms are only for images with global uniform distribution of fog, or can only process a certain type of image, or it is difficult to effectively process images of certain ground features, and the generality of the algorithm is relatively low. Difference
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
[0037] refer to figure 1 , the DCM-HTM haze removal method applicable to optical remote sensing images provided by the present invention comprises the following steps:
[0038] Step 1: Fog area detection and extraction; by selecting a dark channel window with a size of ω×ω, the dark channel image DCM is calculated for the original image with a width and height (W, H), and then mean-shift smoothing is performed on the DCM and according to The histogram automatically sets the threshold to generate a binary image fog_region, and the area with a gray value of 1 in the binary image fog_region is the fog area;
[0039] Step 2: HTM extraction of the fog thickness image; select a non-overlapping window HTM_W(ω',ω'), ω'∈[3,9] for the obtained dark channel image to obtain the fog thickness image HTM , since the HTM obtained after passing through non-overlapping windows is 1 / ω' of the width and height of the original image, it is then filtered and resampled back to the original image si...
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 specifically relates to a DCM-HTM haze-removing method suitably used for optical remote sensing images. The method comprises the following steps of (1) haze zone detection and extraction; (2) haze thickness image HTM extraction; and (3) haze zone correction. The haze thickness image HTM is acquired by selecting a non-overlapping window HTM_W(omega t, omega t) with a size of omega r* omega t from an acquired dark channel image, wherein, the omega t ranges from 3 to 9. The HTM correction is achieved by calculating a natural object background contribution value BACK through the haze thickness image HTM and the extracted fog-region binary image. Finally, on the basis of the radiation transmission principle, original image correction is performed through an HTM haze removing model inside a detected DCM haze zone. The aim of reducing hazes is achieved on the premise that high-brightness objects are prevented from being corrected in an excessive way by selecting a small window from non-brightness objects inside a haze zone range. The HTM is corrected by calculating the normal zone natural object background contribution value BACK on the basis of the detected haze zone and the haze thickness image HTM. The grey level of the normal zone can be well kept.
Description
technical field [0001] The invention relates to the field of optical remote sensing images, in particular to a haze removal method for optical remote sensing images. Background technique [0002] Affected by atmospheric environmental factors such as clouds and smog, the imaging quality of optical satellite remote sensing images is degraded, and the performance of image interpretation and measurement is reduced, which restricts the further application of images. For example, taking the multi-temporal Landsat8OLI 30-meter resolution optical satellite image of the same field of view released by the USGS official website in 2014 including the Dengfeng inspection field area as an example, there are four months that are seriously affected by clouds throughout the year, namely April, June, July and December; and there are five main months affected by smog, namely April, May, June, July and November, accounting for 50% of the whole year. Therefore, how to effectively eliminate or r...
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): G06T5/00
Inventor 朱彩英周杨徐昕东许继伟李鹏飞
Owner KUNSHAN DIGITAL CITY INFORMATION TECH