Brief Communication : Contrast stretching and histogram smoothness based flood detection

Introduction Conclusions References


Introduction
Flood detection/mapping is desirable in variety of applications like disaster management, risk/damage assessment and rehabilitation process.Flood mapping/monitoring techniques use pre and post Synthetic Aperture Radar (SAR) images to classify undated (non flooded) and inundated (flooded) areas (Kussel et al., 2011;Nazir et al., 2013).Visual interpretation (Chambenoit et al., 2003) requires user's involvement for the identification of flooded areas (which is not always feasible).Semi automatic segmentation based flood detection technique requires empirical seed point selection (Dellepiane et al., 2010).
Thresholding based unsupervised flood monitoring (Moser and Serpico, 2006) does not work under complex environmental conditions (in that case users involvement is required for reliable results) (Pulvirenti et al., 2011).Texture matching based scheme (Zhao et al., 2011) suffers from high computational time and overlapping features.Schumann et al. (2009) have applied different processing steps to generate inundation map, however it suffers from reliable calibration and verification.Complex coherence maps are used for the analysis of SAR data for flood monitoring, however, optical images are required for result verification (Chini et al., 2012).Chain of processing based method (Dellepiane and Angiati, 2012)  A contrast enhancement based three steps approach (identical to Dellepiane and Angiati, 2012) is proposed to improve the visibility of SAR images.The technique is composed of three chains (Adaptive Histogram Clipping (AHC), Histogram Remapping (HR) and Histogram Smoothness (HS)).The chains are applied on the pre and post images for the generation of difference image.Fast ready flood map is then generated using equalized pre, post and difference images.Results are evaluated using different data sets which shows the significance of proposed technique.

Proposed methodology
Let I X (l ,m) be pre, I Y (l ,m) be post and I Z(l ,m) be the difference image, where l ∈ [0, . . ., L− 1] and m ∈ [0, . . ., M − 1]. Figure 1 shows the block diagram of proposed technique.
The histogram of pre image I X is clipped (identical to (Dellepiane and Angiati, 2012), but with a low percentile value i.e. q = 0.40) to obtain I X 1 .Note that, at low percentile values (q < 0.40), required details are removed and at higher percentile values (q > 0.40), unwanted details get more prominent thus degrading the quality.Therefore, we have used q = 0.40 because it preserves the required intensity values which contribute to flooding.After AHC, the image I X 1 is then remapped to original intensity range [0-255] using simple linear scaling (Dellepiane and Angiati, 2012) to obtain I X 2 .
HS is then applied to improve the visualization by preserving the details.In contrast to Dellepiane and Angiati (2012), which uses simple Histogram Equalization (HE), we have used HS to maintain the natural look, suppressing unwanted artifacts and enhancing the desired details.
In HS, a smoothness constraint is added to remove abrupt changes using backward difference K (Arici and Dikbas, 2009).The principle is to minimize the difference between modified h X 2 m and current histogram h 2 of image I X 2 such that the modified histogram is also closer to the uniform histogram h X 2 u along with the additional penalty Introduction

Conclusions References
Tables Figures

Back Close
Full (1) The solution to above constraint problem is (Arici and Dikbas, 2009), where, I is identity matrix, α, β are the contrast enhancement and smoothness parameters (chosen empirically as 0.5 and 1000 respectively).The histogram h X 2 m is then used as a mapping function for HE to generate image I X 3 .All three steps (AHC, HR and HS) are applied on I Y (post flooded image) to generate I Y 3 .The difference image I Z is generated as, The images I X and I Y are passed through only CE step to produce ÎX and ÎY respectively.The reason for skipping the first two steps is to preserve intensity values of pre and post images.The purpose of using processed pre and post flooded images for difference image generation is to remove the intensities which contribute very low in flooded areas.
Finally I Z , ÎX and ÎY are combined (by assigning red, blue and green bands) to generate fast ready map.The level of red color is high for pixels whose pre value dominates and vice versa.In RGB image, medium to dark red color represents permanent water like rivers and dark blue color represents the flooded areas.

Simulation and results
Existing and proposed techniques are evaluated on different SAR images."Daichi" (ALOS) on 29 April and 30 July 2006 respectively.Figure 2c and d show the difference images obtained using Dellepiane and Angiati (2012) and proposed technique respectively.In Fig. 2c, the ground details are more prominent while Fig. 2d only highlights the major required details comparatively.The differences in detail contribute significantly in their respective RGB image (shown in Fig. 2e and f).In Fig. 2e, a very high contribution of irrelevant details (of difference image) is visible (for instance, the blue color at the center and at the top right corner). Figure 2f provides better visibility of flooded areas around river (at the top center) and low flooded areas (at the center (below river) and top right corner) of image.Figure 3a and b show the images of Tomakomai, Japan, acquired by Phased Array Type L-band SAR (PALSAR) using H/V and V/V polarization on 19 August 2006 respectively.Figure 3c is the RGB flood map generated using Dellepiane and Angiati (2012) technique.The flood map (in Fig. 3c) highlights some irrelevant details which contribute to flooding (blue colored areas at the right center of image).Figure 3d shows the flood map generated using proposed technique.Figure 3d preserves the natural effect of image as compared to Fig. 3c.

Conclusions
A technique for flood detection based on contrast stretching and histogram smoothing is presented.Different processing steps based on contrast stretching and histogram smoothness are applied on pre, post and difference images to generate flood maps.
Simulation results show improved visualization by maintaining the natural smoothness.

NHESSD Introduction Conclusions References
Tables Figures

Back Close
Full where, I is identity matrix, α, β are the contrast enhancement and smoothness parameters (chosen empirically as 0.5 and sometimes highlight unnecessary details in flood map image.Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Figure 2a and b shows pre and post flood images of ChoeleChoele City, Argentina observed by Discussion Paper | Discussion Paper | Discussion Paper |

Figure 1 .
Figure 1.Flow chart of proposed algorithm.

FigFig. 2 :
Fig. 1: Flow chart of proposed algorithm mapping function for HE to generate image I X3 .All three steps (AHC, HR and HS) are applied on I Y (post flooded image) to generate I Y 3 .The difference image I Z is generated as,I Z (l, m) = 128 + I X3 (l, m) − I Y 3 (l, m) 2 (3) 85The images I X and I Y are passed through only CE step to produce ÎX and ÎY respectively.The reason for skipping the first two steps is to preserve intensity values of pre and post images.The purpose of using processed pre and post flooded images for difference image generation is to remove 90 the intensities which contribute very low in flooded areas.Finally I Z , ÎX and ÎY are combined (by assigning blue, green and red bands) to generate fast ready map.The level of red color is high for pixels whose pre value dominates

Figure 3 .
Figure 3. Evaluation of results using images of Tomakomai, Japan.(a) Pre image acquired on 19 August 2006.(b) Post image acquired on 19 August 2006.(c) Fast ready map generated using Dellepiane and Angiati (2012) technique.(d) Fast ready map generated using proposed technique.