A Newly Developed Algorithm for Cloud Shadow Detection—TIP Method

Abstract

The masking of cloud shadows in optical satellite imagery is an important step in automated processing chains. A new method (the TIP method) for cloud shadow detection in multi-spectral satellite images is presented and compared to current methods. The TIP method is based on the evaluation of thresholds, indices and projections. Most state-of-the-art methods solemnly rely on one of these evaluation steps or on a complex working mechanism. Instead, the new method incorporates three basic evaluation steps into one algorithm for easy and accurate cloud shadow detection. Furthermore the performance of the masking algorithms provided by the software packages ATCOR (“Atmospheric Correction”) and PACO (“Python-based Atmospheric Correction”) is compared with that of the newly implemented TIP method on a set of 20 Sentinel-2 scenes distributed over the globe, covering a wide variety of environments and climates. The algorithms incorporated in each piece of masking software use the class of cloud shadows, but they employ different rules and class-specific thresholds. Classification results are compared to the assessment of an expert human interpreter. The class assignment of the human interpreter is considered as reference or “truth”. The overall accuracies for the class cloud shadows of ATCOR and PACO (including TIP) for difference areas of the selected scenes are 70.4% and 76.6% respectively. The difference area encompasses the parts of the classification image where the classification maps disagree. User and producer accuracies for the class cloud shadow are strongly scene-dependent, typically varying between 45% and 95%. The experimental results show that the proposed TIP method based on thresholds, indices and projections can obtain improved cloud shadow detection performance.

Cloud shadow removal for optical satellite data

“…It works for multispectral and hyperspectral imagery over land acquired by satellite/airborne sensors. Since the deshadowing algorithm relies on the TIP method after, 29 it must be noted that it is applicable for VNIR-SWIR sensors, such as EnMAP, 37 PRISMA, 38 Landsat-8, 39 and Landsat-9, 40 but for VNIR sensors, such as for example the DESIS sensor, 41 the method can only be implemented partially.…”

“…The detailed evaluation of the TIP cloud shadow map calculation can be found in Ref. 29. Having included a better cloud shadow map will improve the results of the shadow removal algorithm.…”

“…Figure 2 shows a flow chart of the eight main steps performed during the new cloud shadow removal algorithm. The steps that have to be performed are: the calculation of the surface reflectance; the evaluation of the cloud shadow map with the TIP method after; 29 the calculation of the MF vector, which gives the unscaled shadow function, Φ; the calculation of the scaled shadow function, normalΦ*; the calculation of the minimum direct sun fraction, amin and the normalized scaled shadow function, normalΦn; and the final step of the shadow removal.…”

“…The new additions are the cloud shadow map from the TIP method developed in Ref. 29 and a newly defined iteration over the scaled minimum shadow function, amin.…”

“…For the detection of the correct location and geometry of a cloud shadow, the direction of observation is crucial since they represent projections of clouds in an image 1 . In this paper, a mono-temporal cloud shadow detection approach is used as preprocessing step for cloud shadow removal, called thresholds, indices, projections (TIP) method 29 …”