Guided filter and adaptive learning rate based non-uniformity correction algorithm for infrared focal plane array

Rong, Shenghui; Zhou, Huixin; Qin, Hanlin; Lai, Rui; Qian, Kun

doi:10.1016/j.infrared.2016.04.037

Cited by 29 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where ij x and ij y are the actual radiation and observed response of the ( ) , i j detector, ij g and ij o denote the gain noise and offset noise, respectively [32].…”

Section: Scheme Of Fpnrmentioning

confidence: 99%

Fixed pattern noise reduction for infrared images based on cascade residual attention CNN

et al. 2020

View full text Add to dashboard Cite

Existing fixed pattern noise reduction (FPNR) methods are easily affected by the motion state of the scene and working condition of the image sensor, which leads to over smooth effects, ghosting artifacts as well as slow convergence rate. To address these issues, we design an innovative cascade convolution neural network (CNN) model with residual skip connections to realize single frame blind FPNR operation without any parameter tuning. Moreover, a coarse-fine convolution (CF-Conv) unit is introduced to extract complementary features in various scales and fuse them to pick more spatial information. Inspired by the success of the visual attention mechanism, we further propose a particular spatial-channel noise attention unit (SCNAU) to separate the scene details from fixed pattern noise more thoroughly and recover the real scene more accurately.Experimental results on test data demonstrate that the proposed cascade CNN-FPNR method outperforms the existing FPNR methods in both of visual effect and quantitative assessment. affected by the fixed pattern noise (FPN), which is mainly caused by the spatial non-uniform response of individual detectors in the sensor [6][7]. More seriously, spatial FPN generally drifts with time, which makes the problem be more challenging [8][9][10][11]. As a result, the FPN causes a significant decline in imaging quality and decreases the precision for object detection and recognition. To meet this challenge, the cost-effective fixed pattern noise reduction (FPNR) techniques based on signal processing are continually investigated and applied in nearly all the infrared imaging systems.Existing FPNR algorithms are mainly divided into two primary categories: reference-based FPNR (RB-FPNR) and scene-based FPNR (SB-FPNR) [12][13][14]. The RB-FPNR methods remove the FPN according to fixed calibration parameters calculated from the response of blackbody radiation at different temperatures [15]. Unfortunately, such a calibration requires the camera to halt the normal operation and update the calibration parameters due to the inherent temporal drift of detector characteristics [16]. Given this fact, most of the recent researches have focused on developing SB-FPNR methods, such as neural networks (NN) [17], temporal high-pass filter (THPF) [18,19] and constant-statistics (CS) method [20][21]. As for SB-FPNR algorithms, the calibration parameters are iteratively updated by utilizing the information extracted from inter-frame motion, therefore, ghosting artifacts and over smooth effects resulted from the sudden deceleration of scene motion often seriously degrade the noise reduction performance, moreover, the relatively slow convergence process occurred in scene switching is unacceptable for most of the practical applications.In recent years, convolution neural network (CNN) [22] models were explored deeply and applied in various image processing tasks [23], such as image super resolution [24,25], image denoising [26], and sketch synthesis [27][28][29]. To the best of our knowledge, CNN based FPNR m...

show abstract

“…where ij x and ij y are the actual radiation and observed response of the ( ) , i j detector, ij g and ij o denote the gain noise and offset noise, respectively [32].…”

Section: Scheme Of Fpnrmentioning

confidence: 99%

Fixed pattern noise reduction for infrared images based on cascade residual attention CNN

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Jara and Torres [32] presented a method to reduce FPN and ghosting artifacts using an enhanced constant statistics method that incorporates a motion threshold. Sheng-Hui et al [33] developed a neural network SBNUC method that combines a guided filter with an adaptive learning rate that demonstrated a decrease in observed ghosting artifacts. Recently, Kuang et al [34] developed a deep learning approach where they used both image denoising and DSR to eliminate striping noise and recover image resolution.…”

Section: Introductionmentioning

confidence: 99%

A Unified Method for Digital Super-Resolution and Restoration in Infrared Microscopy Imaging

et al. 2019

View full text Add to dashboard Cite

Infrared (IR) imaging systems are known to have a range of sensor and optical limitations that result in degraded imagery. Fixed pattern noise (FPN), resulting from pixel-to-pixel response nonuniformity, is a dominant source of error that manifests in collected imagery through the appearance of temporally and spatially correlated noise patterns that are mixed with each image. Furthermore, finite detector size coupled with imperfect system optics can introduce blurring effects and aliasing, ultimately reducing resolution in acquired images. Here, we propose a unified method to reduce FPN and recover high-frequency image content in IR microscopy images. The proposed method uses regularized nonlocal means to highlight spatial features in the scene while maintaining fine textural image details. We derive an iterative optimization method based upon a gradient descent minimization strategy that applies a Wiener deconvolution in each iteration to estimate the blur artifacts. The method is implemented within an embedded mid-wave IR imaging system for microscopy applications. We demonstrate a reduction in FPN and blurring artifacts, achieving improved image resolution in the reconstructed images that are apparent in recovered details on scene objects.

show abstract

“…It usually makes good use of the minimal energy of the image's gradient in some direction and constructs loss function between the output and desired value to constrain the FPN of IRFPA [12]- [17]. One classical method is combined the least mean square (LMS) error estimation and the neural network model, which including NN-NUC [13], BF-NUC [14], TV-NUC [15]- [16] and GF-NUC [17], and etc. These algorithms iteratively calculate the NUC parameters according to the LMS error between the corrected images and their desired images, which are well-known for their low cost of computation and storage resources.…”

Section: Introductionmentioning

confidence: 99%

A Novel Infrared Focal Plane Non-Uniformity Correction Method Based on Co-Occurrence Filter and Adaptive Learning Rate

Lingxiao

Feng

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Non-uniformity commonly exists in the infrared focal plane, which behaves as the fixed pattern noise (FPN) and seriously affects the image quality of the infrared imaging system. This paper proposed a novel scene-based non-uniformity correction method with a new edge-preserve filter and adaptive learning rate. First, using co-occurrence filter as the desired image estimation, the proposed method removed the FPN while preserving the image details. Then, an adaptive learning rate connected with both temporal motion and spatial correlation factor is utilized to decrease the effect of ghosting artifacts. In this way, the proposed method overcomes the shortcomings of the traditional scene-based non-uniformity. Several real infrared image sequences collected in different conditions are used to verify the performance of the proposed method. The experimental results demonstrate that the proposed method has a much better visual effect, making a great balance between the non-uniformity correction and details preservation. Compared with other good NUC methods, this method also has better performance in the aspects of applicability and robustness, which has great application value.INDEX TERMS Non-uniformity correction, IRFPA, edge-preserve filter, adaptive learning rate, ghosting artifacts, details preservation.

show abstract

Guided filter and adaptive learning rate based non-uniformity correction algorithm for infrared focal plane array

Cited by 29 publications

References 11 publications

Fixed pattern noise reduction for infrared images based on cascade residual attention CNN

Fixed pattern noise reduction for infrared images based on cascade residual attention CNN

A Unified Method for Digital Super-Resolution and Restoration in Infrared Microscopy Imaging

A Novel Infrared Focal Plane Non-Uniformity Correction Method Based on Co-Occurrence Filter and Adaptive Learning Rate

Contact Info

Product

Resources

About