A cloud-removal algorithm for SSM/I data
Summary (2 min read)
Introduction
- Because the surface brightness observed by the SSM/I may be adversely affected by spatial variations in the atmospheric profile over the surface, algorithms for cloud removal have been developed [1], [10].
- The authors compare several new algorithms that generate cloud-free composite images from multiple passes of the study region.
- Simulations to determine the effectiveness of these algorithms are performed.
- Section IV introduces the modified maximum average (MMA) and hybrid Manuscript received November 18, 1996; revised October 27, 1997.
II. BACKGROUND
- The SSM/I is a total-power, seven-channel, four-frequency radiometer [5].
- It utilizes an integrate-and-dump filter as the antenna scans the ground track [7].
- The 3-dB antenna footprints, which are different for each frequency, generally have an elliptical shape on the surface of the earth due to the elevation angle of the radiometer [6].
- The most crucial factors affecting a radiometric measurement, however, are the surface emissivity and temperature, the vegetation canopy, and the atmospheric conditions [19].
III. GENERATION OF CLOUD-FREE IMAGES
- One of the challenges in mapping the surface brightness from spaceborne radiometer data is atmospheric distortion.
- Authorized licensed use limited to: Brigham Young University.
- Clouds and precipitation affect surface brightness measurements in two ways.
- Since each frequency has a different footprint size, using lower frequency data to remove atmospheric distortion effects in higher frequency channels may unnecessarily exclude undistorted values in the higher frequency channels [1].
- Thus, the second-highest value technique’s ability to reduce noise is strongly influenced by the measurement distribution.
IV. MMA ALGORITHM
- This algorithm attempts to estimate the cloud-free surface brightness of a pixel by choosing a subset of pixel values from the ensemble of measurements of that pixel and then averaging the selected values together.
- To select pixel values from the ensemble in the MMA technique, the sample mean of the entire pixel ensemble is first computed.
- The remaining values consist of those values that are above the ensemble mean but less than the maximum value of the ensemble.
- To compare the variances of the MMA algorithm and the second-highest value technique, consider Fig. 2(b).
- In the standard deviation image of Fig. 4, the areas with low values correspond to regions with little or no atmospheric distortion.
V. SIMULATION
- To further compare and contrast the mean, second-highest value, MMA, and hybrid algorithms, a simple Monte Carlo analysis for a single pixel is presented.
- This simulation assumes that the true pixel brightness for a geographical area is 280 K.
- Two of the ensemble measurements then have simulated atmospheric distortion added.
- The windowed mean is the mean of values within one standard deviation of the ensemble mean.
- The simulation results indicate that MMA is superior in the presence of significant distortion and mean is best with little or no distortion present, while the hybrid algorithm combines the two in a manner that uses the appropriate algorithm for each pixel.
VI. SSM/I DATA ANALYSIS
- To validate the algorithms with actual data, a region of the Amazon Basin was chosen for SSM/I data analysis.
- Authorized licensed use limited to: Brigham Young University.
- These distributions demonstrate the difference between cloudy and clear regions.
- The authors note that the mean image has the highest standard deviation.
- Their similar performance indicates that hybrid primarily used the MMA algorithm in the presence of atmospheric distortion.
VII. CONCLUSION
- A comparison of several different methods (mean, secondhighest value, MMA, and hybrid) for creating cloud-free temporal composite surface brightness temperature images from SSM/I has been presented.
- Taking the mean is optimum in the absence of clouds and hydrometeors.
- The second-highest value algorithm removes distorted pixel brightness temperature measurements, but it has an inherent high bias in its brightness temperature distribution.
- The MMA algorithm more accurately estimates the desired value and has a lower variance; however, it may have an undesirable bias for instances in which no distortion is present.
- The authors note that in the presence of persistent atmospheric distortions, the distortion cannot be removed.
Did you find this useful? Give us your feedback
Citations
132 citations
Cites background or methods from "A cloud-removal algorithm for SSM/I..."
...Selection of the MMA subset of measurements is carried out by considering all the measurements above the vector mean except the one with the highest value....
[...]
...In [25], the authors show that a hybrid algorithm that implements MMA in the presence of clouds and averages the measurements in their absence, can significantly improve the quality of the composite image compared to the MMA and the SH algorithms....
[...]
...For the purpose of comparison, we implemented two other methods based on compositing algorithms developed to reconstruct cloudy areas in images acquired by the microwave SSM/I sensor [25]....
[...]
...Several algorithms, like the second highest (SH) [24] and the modified maximum average (MMA) [25], have been developed to remove cloud effects from SSM/I images....
[...]
131 citations
Cites background or methods from "A cloud-removal algorithm for SSM/I..."
...[5] for recovering Advanced Very High Resolution Radiometer measurements that are modified by the effects not only of clouds but also of cloud shadows....
[...]
...Each of the obscured region is reconstructed using our bandelet-based method as well as two other multitemporal analysis methods, namely, the modified maximum average [5] and the nearest neighbor method....
[...]
56 citations
Cites methods from "A cloud-removal algorithm for SSM/I..."
...Among the relatively few works available in the cloud removal literature, one can find the compositing approaches [3], a cloud-specific retrieval algorithm [4], a method based on regression trees and histogram matching [5], and an ecosystem classification-dependent temporal interpolation technique [6]....
[...]
32 citations
Cites background from "A cloud-removal algorithm for SSM/I..."
...The algorithm for the generation of the composite brightness temperature product is discussed in Long et al. (1999)....
[...]
31 citations
Cites background or methods from "A cloud-removal algorithm for SSM/I..."
...lected from previous time periods with spatial relationships, and are mainly applied to the homogeneous landscapes with less dynamic nature [6]–[9]....
[...]
...Other possible methods, such as the second highest value [48] and the modified maximum average [6], could be utilized to optimize the multiple predicted results for possible...
[...]
References
57 citations
"A cloud-removal algorithm for SSM/I..." refers methods in this paper
...SSM/I data have also been used for land and ice studies, including measurements of snow cover classification [4], soil and plant moisture content [8], [15], atmospheric moisture over land [ 10 ], land surface temperature [12], and mapping polar ice [18]....
[...]
...Because the surface brightness observed by the SSM/I may be adversely affected by spatial variations in the atmospheric profile over the surface, algorithms for cloud removal have been developed [1], [ 10 ]....
[...]
40 citations
39 citations
"A cloud-removal algorithm for SSM/I..." refers methods in this paper
...SSM/I data have also been used for land and ice studies, including measurements of snow cover classification [4], soil and plant moisture content [8], [15], atmospheric moisture over land [10], land surface temperature [12], and mapping polar ice [18]....
[...]
15 citations