Aerial image dehazing using a deep convolutional autoencoder
- PDF / 5,210,867 Bytes
- 19 Pages / 439.642 x 666.49 pts Page_size
- 39 Downloads / 228 Views
Aerial image dehazing using a deep convolutional autoencoder Hamidreza Fazlali1 Thia Kirubarajan1
· Shahram Shirani1 · Mike McDonald1 · Daly Brown1 ·
Received: 7 October 2019 / Revised: 6 July 2020 / Accepted: 16 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Aerial images and videos are extensively used for object detection and target tracking. However, due to the presence of thin clouds, haze or smoke from buildings, the processing of aerial data can be challenging. Existing single-image dehazing methods that work on ground-to-ground images, do not perform well on aerial images. Moreover, current dehazing methods are not capable for real-time processing. In this paper, a new end-to-end aerial image dehazing method using a deep convolutional autoencoder is proposed. Using the convolutional autoencoder, the dehazing problem is divided into two parts, namely, encoder, which aims extract important features to dehaze hazy regions and decoder, which aims to reconstruct the dehazed image using the down-sampled image received from the encoder. In this proposed method, we also exploit the superpixels in two different scales to generate synthetic thin cloud data to train our network. Since this network is trained in an end-to-end manner, in the test phase, for each input hazy aerial image, the proposed algorithm outputs a dehazed version without requiring any other information such as transmission map or atmospheric light value. With the proposed method, hazy regions are dehazed and objects within hazy regions become more visible while the contrast of non-hazy regions is increased. Experimental results on synthetic and real hazy aerial images demonstrate the superiority of the proposed method compared to existing dehazing methods in terms of quality and speed. Keywords Aerial images and videos · Dehazing · Deep convolutional autoencoder · Airborne surveillance · Wide area motion imagery
1 Introduction Aerial images or videos captured by cameras on airplanes or Unmanned Aerial Vehicles (UAVs) are often used for detection and tracking of ground objects. Such data are collected at low altitudes compared to satellite-based data. Several methods have been proposed in the Hamidreza Fazlali
[email protected] 1
ECE Department of McMaster University, 1280 Main Street West, Hamilton, ON, Canada
Multimedia Tools and Applications
literature to detect and track moving vehicles in airborne videos [20, 25, 29, 30]. However, these methods do not usually consider challenges such as hazy conditions with thin clouds or smoke from buildings. In such scenarios, small particles floating in the atmosphere can scatter the light reflected back to the camera, reducing the intensity of reflected signals and degrading the visual quality of the image. Moreover, due to the light scattering phenomenon, the contrast of the received image decreases and the colors therein become faint. Therefore, haze removal is a key pre-processing step for any computer vision algorithm for aerial images. Many meth
Data Loading...
