CN105279738A - Coping method of shadow problem in vegetation parameter estimation, based on remote sensing images - Google Patents

Abstract

The invention discloses a method for coping with a shadow problem caused by landform fluctuating and cloud clusters in vegetation parameter estimation, based on remote sensing images. The method comprises the steps of: firstly carrying out atmosphere correction on the remote sensing images, then selecting spectrum wave bands sensitive to target parameters, and finally taking the spectrum wave bands to a correction formula y=(b1-b2)/(b1+b2) so as to obtain normalized vegetation indexes after final correction. The adopted correction formula is simple linear operation of two wave bands of remote sensing data, the complexity is low, the calculation speed is substantially improved, and the batch processing of remote sensing images is facilitated. From a correction result, the effect is substantial, the shadow problem is relieved to a certain degree, and the sensitivity to the target parameters is maintained, so that the technical method provided by the invention for relieving the shadow problem caused by landform fluctuating and cloud clusters is simple and applicable.

Description

Countermeasures for Shadow Problems in Vegetation Parameter Estimation Based on Remote Sensing Images

technical field

The invention belongs to the technical field of remote sensing, and relates to a method for dealing with shadow problems in vegetation parameter estimation of remote sensing images.

Background technique

Vegetation biochemical-biophysical parameters, such as leaf area index, chlorophyll, leaf water content, canopy water content, biomass, etc., respond to vegetation growth conditions, estimate material and energy exchange rates between the lower atmosphere and vegetation, and understand crop water stress It plays an important role in indicating the extent of forest and grassland fire risks and assessing the probability of occurrence of forest and grassland fire risks. These important vegetation parameters have different degrees of sensitivity in different spectral bands of vegetation. For example, the leaf area index shows strong reflection characteristics in the near-infrared band, and chlorophyll shows strong absorption properties in the red band and strong reflection in the green band. The water content of the canopy has strong absorption properties in the short-wave infrared band. At the same time, remote sensing image data also has the characteristics of multi-temporal, multi-spatial resolution, and timeliness, so that the data can be used to quickly extract target parameters in large-scale space and multi-temporal phases, and improve scientific and timely decision-making information for relevant industries. However, terrain undulations and shadows formed by some clouds will lead to spatial discontinuity in remote sensing images or get wrong target parameter estimates, hindering the further application of remote sensing technology.

Contents of the invention

The purpose of the present invention is to provide a simple and quick method for dealing with terrain fluctuations and shadows caused by clouds in vegetation parameter estimation based on remote sensing image data, so as to alleviate the problems of remote sensing image space discontinuity and target parameter misestimation caused by shadows.

The technical scheme adopted by the present invention to solve the above-mentioned technical problems is: firstly, perform atmospheric correction on the remote sensing image, and convert the DN value into the surface reflectance; secondly, select the spectral band sensitive to the target vegetation parameters (absorption or reflection characteristics); finally , bring the reflectance of any two bands in the above step into the following formula, y=(b1-b2)/(b1+b2), where b1 and b2 are any two non-repetitive bands sensitive to vegetation, specifically green light band (500-600nm), red band (600-700nm), near-infrared band (760-900nm) and short-wave infrared band (1500-2400nm), y is the final normalized index, and its value range It is between -1 and 1. The final index obtained by using this technical scheme can not only ensure its sensitivity to the target vegetation parameters, but also alleviate the shadow problems caused by terrain fluctuations and clouds to a certain extent.

Furthermore, remote sensing images specifically refer to remote sensing image data of areas covered by vegetation.

Further, b1 and b2 must be atmospherically corrected surface albedo image data.

Further, b1 and b2 must be spectral bands sensitive to target parameters (absorption or reflection characteristics), specifically green light band (500-600nm), red light band (600-700nm), near-infrared band (760-900nm) ) and short-wave infrared band (1500-2400nm).

Further, the value range of y is between -1 and 1.

The beneficial effect of the present invention is that even if the original remote sensing image data is subjected to atmospheric correction, the problem of shadows caused by terrain fluctuations and clouds cannot be alleviated. The final normalization index obtained by the method provided by the present invention can not only ensure its sensitivity to target parameters, but also alleviate the influence of shadows to a certain extent. At the same time, the method is simple, fast, and easy to operate, and can be used to alleviate the shadow problem of remote sensing images in a large scale, ensure the spatial continuity of remote sensing image data and improve the estimation accuracy of target parameters as a whole.

Description of drawings

Accompanying drawing 1 is the comparison effect of the original image and the remote sensing image corrected by this patent method

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Figure 1(a) is a real-color synthetic image of Landsat8. It can be seen from the figure that the shadow problem caused by mountains and clouds is very serious. If the image data is directly used to estimate the target parameters, it will undoubtedly lead to spatial discontinuity in the remote sensing image of the study area or more serious parameter misestimation problems. . The method of this patent is now used for the following processing: firstly, the atmospheric correction is performed on the image to convert the DN value into the apparent reflectance. The commonly used atmospheric correction software includes the FLAASH module of ENVI and the 6S atmospheric correction model. The input parameters of the two are similar, and can be obtained from the metadata file of the remote sensing data and the data released by the local weather station. Second, select the target band. Generally, red band (band4), near-infrared band (band5) and short-wave infrared band (band6) are used to estimate vegetation parameters. In this example, the three bands are also selected as the calibration object. Finally, the selected band image data is brought into y=(b1-b2)/(b1+b2) to obtain the final normalized index. As shown in Figure 1(b) and (c), after correction, the shadows in the remote sensing images are greatly relieved, especially the relief of shadows in (b) is more obvious; at the same time, this method guarantees the correction High sensitivity of the afterimage to vegetation. It can be seen that using this patent can effectively alleviate the shadow problems caused by terrain fluctuations and clouds.

The correction formula used in the method for alleviating shadows on remote sensing images in this patent is a simple linear operation of two bands of remote sensing data, with low complexity, which can greatly increase the calculation rate and facilitate batch processing of remote sensing images. Judging from the correction results, the effect is remarkable, which alleviates the shadow problem to a certain extent and maintains the sensitivity to the target parameters. Therefore, the technical method provided by this patent is simple and practical for alleviating the shadow problems caused by terrain fluctuations and clouds.

Claims (6)

1., based on the countermeasure of shadow problem in the estimation of remote sensing image vegetation parameter, it is characterized in that comprising the following steps: first, atmospheric correction is done to remote sensing image, DN value is converted to Reflectivity for Growing Season; Secondly, the spectral band to target vegetation parameter sensitivity (absorbing or reflection characteristic) is chosen; Finally, two wave band reflectivity any in upper step are brought into following formula, y=(b1-b2)/(b1+b2), wherein, b1 is any two and does not repeat wave band, and y is the index after the normalization finally obtained, and its span is between-1 to 1.

2. as claimed in claim 1, it is characterized in that: remote sensing image refers in particular to the remote sensing image data in vegetative coverage region.

3. as claimed in claim 1, it is characterized in that: updating formula used is y=(b1-b2)/(b1+b2).

4. as claimed in claim 1, it is characterized in that b1 and b2 must for the Reflectivity for Growing Season image data after atmospheric correction.

5. as claimed in claim 1, it is characterized in that: b1 and b2 for the spectral band to target vegetation parameter sensitivity (absorbing or reflection characteristic), must be specially green light band (500-600nm), red spectral band (600-700nm), near-infrared band (760-900nm) and short infrared wave band (1500-2400nm).

6. as claimed in claim 1, it is characterized in that: the span of y is between-1 to 1.