Conference Paper

An automatic wavelet-based nonlinear image enhancement technique for aerial imagery

Turkish Air Force Acad., Istanbul, Turkey
DOI: 10.1109/RAST.2009.5158217 Conference: Recent Advances in Space Technologies, 2009. RAST '09. 4th International Conference on
Source: IEEE Xplore


Recently we proposed a wavelet-based dynamic range compression algorithm to improve the visual quality of digital images captured in the high dynamic range scenes with non-uniform lighting conditions. The fast image enhancement algorithm which provides dynamic range compression preserving the local contrast and tonal rendition is a very good candidate in aerial imagery applications such as image interpretation for defense and security tasks. This algorithm can further be applied to video streaming for aviation safety. In this paper the latest version of the proposed algorithm which is able to enhance aerial images so that the enhanced images are better then direct human observation, is presented. The results obtained by applying the algorithm to numerous aerial images show strong robustness and high image quality.

Download full-text


Available from: Numan Unaldi, Jan 31, 2015
  • Source
    • "The airlight map was generated by modeling the relationship between coordinates of image pixels and regional airlight. Unaldi et al. [12] provided a wavelet-based dynamic range compression algorithm to enhancement aerial imagery. They used histogram adjustment for minimizing the illumination effect, wavelet based dynamic range compression for enhancing local contrast, and a non-linear approach for restoring color. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Turbidity such as fog, mist, haze, and smoke progressively reduces image contrast and visibility with increasing distance. In this paper, we propose an algorithm to degrade the turbid degree from the turbid image. The turbidity can be considered as a kind of noise. The basic assumption of the proposed algorithm is that an image consists of a reference intensity level and a characteristic intensity level. The reference intensity level is considered as general or background intensity level and it can be obtained by a low pass filter. The characteristic intensity level can be calculated by subtracting the reference intensity level from the original intensity level in the given image. The human eye has a logarithmic intensity response so the target intensity level will be created by the adaptive logarithmic function to approach the human vision. The turbid image will be degraded by transforming the characteristic intensity level into the target intensity level according to the proportion of the reference intensity level to the chosen target intensity level. The experimental results show the varied degraded turbid image as well as compared with other algorithms.
    Full-text · Article · Sep 2012 · Computers & Mathematics with Applications
  • [Show abstract] [Hide abstract]
    ABSTRACT: Most of the techniques developed for infrared (IR) image enhancement (IE) depend heavily on the scene, environmental conditions, and the properties of the imaging system. So, with a set of predefined scenario properties, a content-based IR-IE technique can be developed for better situational awareness. This study proposes an adaptive IR-IE technique based on clustering of wavelet coefficients of an image for sea surveillance systems. Discrete wavelet transform (DWT) of an image is computed and feature vectors are constructed from subband images. Clustering operation is applied to group similar feature vectors that belong to different scene components such as target or background. Depending on the feature vectors, a weight is assigned to each cluster and these weights are used to compute gain matrices which are used to multiply wavelet coefficients for the enhancement of the original image. Enhancement results are presented and a comparison of the performance of the proposed algorithm is given through subjective tests with other well known frequency and histogram based enhancement techniques. The proposed algorithm outperforms previous ones in the truthfulness, detail visibility of the target, artificiality, and total quality criteria, while providing an acceptable computational load.Highlights► Decompose the original infrared sea-surface image by discrete wavelet transform. ► Cluster wavelet coefficients to group similar feature vectors as target or background. ► Assign weights to each cluster and compute gain matrices. ► Multiply each subband image with the appropriate gain matrix. ► Reconstruct the new wavelet coefficients by inverse discrete wavelet transform to obtain enhanced image.
    No preview · Article · Sep 2011 · Infrared Physics & Technology
  • [Show abstract] [Hide abstract]
    ABSTRACT: 14-bit or 16-bit pixel depth high dynamic-range images are acquired from visible band cameras and from infrared imaging devices which are more widely used nowadays. Usually, linear mapping is used to display these images to operators. However, results of the researches done to map images into 0-255 range in recent years show that different techniques result in major differences at image perception and detail visibility. Successful compression of image dynamic range increases the operator awareness for surveillance systems and ensure more effective display of scene details to user. Besides, dynamic-range compression techniques effect the enhancement of the success rate of image target detection and tracking techniques. In this work, scene components are analyzed using wavelet coefficients and intensity distribution of scene components are extracted. Extracted intensity distribution is used to display scene components effectively.
    No preview · Conference Paper · Jan 2013