High Precision Restoration Method for Non-uniformly Warped Images

Halder, Kalyan Kumar; Tahtalı, Murat; Anavatti, Sreenatha G.

doi:10.1007/978-3-319-02895-8_6

Cited by 6 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this method, the first frame is considered as the reference frame. A backward mapping [29] non-rigid image registration is employed to estimate the pixel shift maps x s x; y; t and y s x; y; t of all other frames with respect to the reference frame. By using the shift maps, therefore, the centroids C x and C y are calculated by averaging…”

Section: Proposed Image Restoration Methodsmentioning

confidence: 99%

Simple and efficient approach for restoration of non-uniformly warped images

2014

Self Cite

View full text Add to dashboard Cite

A high accuracy image dewarping method is proposed to restore images from non-uniformly warped video sequences degraded by atmospheric turbulence. This approach contains three major steps. First, a non-rigid image registration technique is employed to register all the frames in the sequence to a reference frame and estimate the motion fields. Second, an iterative First Register Then Average And Subtract (iFRTAAS) method is applied to correct the geometric deformations of the warped frames. The third step involves applying a non-local means filter for the compensation of noise and to improve the signal-to-noise ratio (SNR) of the restored reference frame. Simulations are carried out by applying the method to synthetic and real-life turbulence degraded videos and by determining various quality metrics. A performance comparison is presented between the proposed method and two earlier methods, which verifies that the proposed method provides significant improvement on the image restoration accuracy.

show abstract

Section: Proposed Image Restoration Methodsmentioning

confidence: 99%

Simple and efficient approach for restoration of non-uniformly warped images

2014

Self Cite

View full text Add to dashboard Cite

show abstract

“…There are a fairly large number of methods in the literature that address the problem of restoring nonuniformly warped images degraded by atmospheric turbulence [4][5][6][7][8][9][10][11]. These methods could further be enhanced if the turbulence-induced warping can be predicted ahead of time.…”

Section: Introductionmentioning

confidence: 99%

“…These degradation effects are in part due to random inhomogeneities in the temperature distribution, causing variations of refractive index along the optical transmission path which are more prominent near to the ground [1][2][3][4]. Although the dynamics of atmospheric turbulence can be very complicated and diverse, the resulting effects can simply be observed as random geometric distortions and nonuniform blurring [5][6][7][8]. Imaging through the atmospheric turbulence involves the circumvention of these degrading effects through various techniques, such as adaptive optics (AO) and digital image processing.…”

Section: Introductionmentioning

confidence: 99%

Model-free prediction of atmospheric warp based on artificial neural network

2014

Self Cite

View full text Add to dashboard Cite

This paper presents the application of artificial neural network for predicting the warping of images of remote objects or scenes ahead of time. The algorithm is based on estimating the pattern of warping of previously captured short-exposure frames through a generalized regression neural network (GRNN) and then predicting the warping of the upcoming frame. A high-accuracy optical flow technique is employed to estimate the dense motion fields of the captured frames, which are considered as training data for the GRNN. The proposed approach is independent of the pixel-oscillatory model unlike the state-of-the-art Kalman filter (KF) approach. Simulation experiments on synthetic and real-world turbulence degraded videos show that the proposed GRNN-based approach performs better than the KF approach in atmospheric warp prediction.

show abstract

“…Based on the same strategy, Zhu et al [14] used a B‐spline registration within a Bayesian framework, involving a bilateral total variation regularisation, to estimate the inverse operators. A two‐steps method using first a multiscale optical flow estimation and then the first register then average and subtract algorithm was proposed in [15, 16] to obtain a restored image. The authors in [17, 18], respectively, use a generalised regression neural network and a convolutional neural network to learn turbulence‐induced deformations.…”

Section: Introductionmentioning

confidence: 99%