A Combined Stripe Noise Removal and Deblurring Recovering Method for Thermal Infrared Remote Sensing Images

Zhang, Jingwen; Zhou, Xiaoxuan; Li, Liyuan; Hu, Tingliang; Chen, Fansheng

doi:10.1109/tgrs.2022.3196050

Cited by 22 publications

(6 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…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.…”

Section: Methodsmentioning

confidence: 99%

A New GRACE Filtering Approach Based on Iterative Image Convolution

Zhou,

Wang,

Tang

et al. 2023

JGR Solid Earth

View full text Add to dashboard Cite

The Gravity Recovery and Climate Experiment (GRACE) mission and its successor GRACE Follow‐On (GRACE‐FO) mission provide humans with a unique perspective on mass distribution through accurate temporal gravity field models. However, these models suffer from unavoidable stripe errors in the spatial domain. A new approach based on image processing has been developed for filtering GRACE and GRACE‐FO spherical harmonic solutions. In this approach, the gravity fields are treated as noise‐contaminated images, and the stripe errors are expected to be suppressed by convolving the input with an appropriate kernel. To verify the effectiveness of the proposed approach, a closed‐loop simulation environment over the nominal mission lifetime of 5 years is designed. Image processing indicators and geophysical information evaluation results demonstrate the superior performance of the new approach in suppressing stripe errors. In the context of real GRACE data processing, our new approach still yields better results as follows: 1) The comprehensive ability of de‐striping and retaining details has slight advantages over some commonly used filters, as demonstrated through spatial distributions, amplitude spectra, and coefficient spectra. 2) The equivalent water height time series at the basin scale is comparable to the results derived from other filters or Level‐3 products. 3) The estimated signal amplitude of the Mw9.1 2004 Sumatra earthquake co‐seismic gravity change is 52.7% larger than that derived from the Gaussian 300 km filter, which is much closer to the theoretical value derived from the dislocation theory.

show abstract

Section: Methodsmentioning

confidence: 99%

A New GRACE Filtering Approach Based on Iterative Image Convolution

Zhou,

Wang,

Tang

et al. 2023

JGR Solid Earth

View full text Add to dashboard Cite

show abstract

“…13 reveals a pronounced correlation between the optimal 𝜖𝜖 values and the specific intervals of stripe intensities. More specifically, within the intensity ranges of [0-2], [2][3][4][5][6], [6][7][8][9][10], [10][11][12][13][14][15], and [15][16][17] as the 𝜖𝜖 values were sequentially set to 1, 3, 5, 10, 20, the resultant images achieved the best PSNR values. Based on these findings, we have selected [2,6,10,15] as our reference points for threshold values, applying these in both our simulation experiments and real-image testing.…”

Section: Parameter Analysismentioning

confidence: 99%

“…are arranged in rows or columns [2][3][4][5][6]. The existence of nonuniformity noise has detrimental effects on image quality, diminishes the signal-to-noise ratio, and significantly influences subsequent processing tasks, such as target detection.…”

mentioning

confidence: 99%

“…The objective function is iteratively minimized to achieve the desired outcome of reducing stripe intensity or eliminating stripes. These methods can be categorized into three primary classes: full variational methods [17,[27][28][29], low-rank methods [3,5,30,31], and sparse constraint methods [32][33][34][35][36]. Although these methods exhibit strong adaptability and achieve effective destriping results, they are limited by their high computational complexity.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A Weighted Guided Filtering-Based Multidomain Fusion Destriping Method

Hong,

Rao,

Zhou

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

An infrared camera is affected by the photon effect, temperature changes, time drift and other factors when operating in orbit, which makes the ground non-uniformity correction coefficient invalid, resulting in non-uniformity stripes in the infrared images and restricting their practicality in further analysis and applications. Existing destriping methods often suffer from loss of image details and artifact generation. To solve this problem, we proposed a weighted guided filtering-based multidomain fusion destriping approach that leverages the structural, directional, and spectral characteristics of stripe noise. Firstly, we addressed the issue of artifacts caused by Fourier domain filtering through an adaptive filtering approach that employs a variable threshold to minimize filtering induced artifacts and obtain clearer guided images. Furthermore, capitalizing on the directional properties of wavelet decomposition effectively separates image information from stripe information. To integrate the advantages of both approaches, we employed a weighted guided filter to seamlessly fuse the guided image with the wavelet decomposition image. In terms of quantitative metrics, the proposed method generally beats the other five comparative methods, with significant improvements in image PSNR, SSIM, NIQE, and MRD, particularly for complex images where the enhancements were more pronounced. These experimental results collectively demonstrate the significant progress achieved by the proposed method in effectively reducing stripe noise, better preserving the original structural details of the image, and suppressing the occurrence of artifacts.Index Terms-stripe noise; thermal infrared; Fourier transform; wavelet decomposition; image fusion; weighted guided filter; I. INTRODUCTION n recent years, infrared remote sensing technology has made significant advancements, expanding its applications across various fields, such as urban management, environmental protection, and military operations. During orbit operation, however, the detector is affected by temperature, the photon effect, aging, and other factors, which can make the ground non-uniformity correction factor invalid, resulting in non-uniformity in the image. This non-uniformity typically manifests as fixed pattern noise [1]. Currently, this type of stripe noise exists mainly in either horizontal or vertical directions, due to the fact that most of the sensor readout circuits This paragraph of the first footnote will contain the date on which you submitted your paper for review, which is populated by IEEE. It is IEEE style to display support information.

show abstract

“…Data tunggal artinya suhu yang didapatkan dari perangkat ini berupa satu data saja seperti ditemui pada thermo gun. Sedangkan data majemuk dapat ditemui pada perangkat yang luarannya adalah gambar pemetaan suhu objek seperti pada kamera thermal [12][13] [14][15] [16]. Kelemahan pada pembacaan suhu memanfaatkan fenomena infra merah yaitu hasil pembacaan yang tidak stabil dan akurasinya yang jauh jika dibandingkan dengan pengukuran dengan alat ukur yang bersentuhan langsung dengan object yang ingin diukur.…”

Section: Pendahuluanunclassified

Estimasi Galat Sebagai Kompensasi Hasil Pembacaan Sensor Suhu Non-Sentuh Menggunakan Regresi Linier

Purnama¹,

Afandi²,

Selamet³

et al. 2023

View full text Add to dashboard Cite

Abstrak Penelitian ini bertujuan untuk membuktikan penerapan metode regresi linear untuk mengestimasi galat sebagai kompensasi untuk menurunkan galat pembacaan sensor suhu non-sentuh. Hal ini terlihat pada masa pandemi COVID – 19 melalui penggunaan thermo gun untuk mengukur suhu seseorang membutuhkan jarak yang sangat dekat. Penelitian dilakukan dengan mengukur objek air yang merupakan unsur terbanyak pada tubuh manusia. Suhu pengukuran telah dikondisikan pada 36°C, 37°C, dan 38°C dengan masing-masing variasi jarak 2cm, 4cm, dan 6cm. Hasil dari validasi menunjukkan model dapat menurunkan galat pada suhu 36°C pada masing-masing jarak sebesar 4,43%, 8,00%, dan 8,70% pada jarak 2cm, 4cm, dan 6cm. Validasi berdasarkan MAE sebelum dimodelkan adalah 2,91 dan nilai MAE setelah dimodelkan adalah 0,37. Penurunan nilai galat pada suhu 37°C untuk masing-masing jarak 2cm, 4cm, dan 6cm adalah 5,94%, 8,72%, dan 7,34%. Nilai validasi berdasarkan MAE sebelum dimodelkan adalah 3,02 dan nilai MAE setelah dimodelkan adalah 0,27. Penurunan nilai galat pada suhu 38°C untuk masing-masing jarak 2cm, 4cm, dan 6cm adalah 6,27%, 8,67%, dan 11,03%. Nilai validasi berdasarkan MAE sebelum dilakukan permodelan adalah 3,87 dan nilai MAE setelah dimodelkan adalah 0,57. Berdasarkan data tersebut dapat disimpulkan bahwa estimasi galat sebagai kompensasi hasil pembacaan sensor suhu non-sentuh mampu menurunkan galat hasil pembacaan sensor.

show abstract

A Combined Stripe Noise Removal and Deblurring Recovering Method for Thermal Infrared Remote Sensing Images

Cited by 22 publications

References 60 publications

A New GRACE Filtering Approach Based on Iterative Image Convolution

A New GRACE Filtering Approach Based on Iterative Image Convolution

A Weighted Guided Filtering-Based Multidomain Fusion Destriping Method

Estimasi Galat Sebagai Kompensasi Hasil Pembacaan Sensor Suhu Non-Sentuh Menggunakan Regresi Linier

Contact Info

Product

Resources

About