Fourier Domain Anomaly Detection and Spectral Fusion for Stripe Noise Removal of TIR Imagery

Abstract

Stripe noise is a common and unwelcome noise pattern in various thermal infrared (TIR) image data including conventional TIR images and remote sensing TIR spectral images. Most existing stripe noise removal (destriping) methods are often difficult to keep a good and robust efficacy in dealing with the real-life complex noise cases. In this paper, based on the intrinsic spectral properties of TIR images and stripe noise, we propose a novel two-stage transform domain destriping method called Fourier domain anomaly detection and spectral fusion (ADSF). Considering the principal frequencies polluted by stripe noise as outliers in the statistical spectrum of TIR images, our naive idea is first to detect the potential anomalies and then correct them effectively in the Fourier domain to reconstruct a desired destriping result. More specifically, anomaly detection for stripe frequencies is achieved through a regional comparison between the original spectrum and the expected spectrum that statistically follows a generalized Laplacian regression model, and then an anomaly weight map is generated accordingly. In the correction stage, we propose a guidance-image-based spectrum fusion strategy, which integrates the original spectrum and the spectrum of a guidance image via the anomaly weight map. The final reconstruction result not only has no stripe noise but also maintains image structures and details well. Extensive real experiments are performed on conventional TIR images and remote sensing spectral images, respectively. The qualitative and quantitative assessment results demonstrate the superior effectiveness and strong robustness of the proposed method.

A New GRACE Filtering Approach Based on Iterative Image Convolution

“…As mentioned in the introduction, some repetitive texture structures in the image can be suppressed by adjusting their frequency spectrum. This kind of strategy is commonly applied in image processing of remote sensing (Jinsong et al., 2003; Zeng et al., 2020; Zhang et al., 2022). Considering that the stripes are not strictly periodic, and the adjustment of the frequency spectrum often shows the oscillating Gibbs effect in the space, we have designed an iterative method that combines convolution with Gaussian smoothing.…”

An Infrared Stripe Noise Removal Method Based on Multi-Scale Wavelet Transform and Multinomial Sparse Representation

“…The processing results of each algorithm need to be evaluated. There are two methods of evaluation: one is subjective evaluation, which is mainly through the human eye to observe the degree of stripe noise removal and the level of information protection in non-stripe noise regions in the image; the other is objective indexes, and three evaluation indexes that are widely recognized by researchers in this field are selected in this section, namely: NR (noise reduction), which reflects the denoising ability of the algorithm [ 28 , 29 ], MRD (mean relative deviation), which characterizes the degree of information loss in the non-stripe noise region of the image after denoising [ 29 , 30 ], and ID (image distortion), which reflects the fidelity level of the algorithm [ 15 , 31 ]. The definitions of these three metrics are shown in formulas ( 32 )–( 34 ).…”

Wavelet Channel Attention Module With A Fusion Network For Single Image Deraining

“…Quantitative Results: The quantitative results are shown in Table I. We added Gaussian-distributed stripe noise with noise level σ ∈ [0, 0.10], Uniform-distributed stripe noise with noise level µ ∈ [−0.10, 0.10], and periodic stripe noise T ∈ [6,7,8,9] on each test sets. The best, second-best, and third-best results are indicated in bold, underlined, and italicized font, respectively.…”

“…Due to the widespread applications of the task, numerous techniques have been proposed for single-image destriping. For instance, various methods use handcrafted features to distinguish between background and stripe components of the IR image, which employ filters [8], [9], [10], data statistics [11], [12], [13], and other optimization techniques [14], [15], [16]. They leverage empirical observations of image properties that can achieve acceptable performance for relatively simple scenes.…”