Scratch detection supported by coherency analysis of motion vector fields

Buhler, Jurgen; Weitbruch, S.; Thébault, Cédric; Doser, Ingo; Neisse, Olaf

doi:10.1109/icip.2009.5414090

Cited by 13 publications

(16 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [19], a multiple hypothesis tracker (MHT) is utilized for scratch tracking. Gulu et al [20] suggest a block-matching manner in motion compensated frames to distinguish scratches with a predefined threshold, and Muller et al [21] take into account precise movement information as well as a hierarchical block-matching algorithm optimized toward tracking vertical objects. These detection methods, though achieving better accuracy than spatial-only detectors, are likely to be confused by complicated real scratch paths along time axis, which pose difficulty and instability on tracking or block-matching mechanism, and the huge computational load has hampered their widespread use.…”

Section: A Prior Workmentioning

confidence: 99%

“…As shown in Figs. 2 and 3, we change the value of μ from 0.8 to 4 and plot the corresponding function curves of: true-positive rate R TP 1 (μ) in (18) [R TP 2 (μ) in (20)] and false-positive rate R FP 1 (μ) in (19) [R FP 2 (μ) in (21)] for the eight synthetic videos. We also set the R FP 1 (μ) ( R FP 2 (μ)) as a horizontal ordinate and the R TP 1 (μ) (R TP 2 (μ)) as a vertical ordinate to plot the receiver operating characteristic (ROC) curves to evaluate the decomposition performance of different μ.…”

Section: B Parameter Settingmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Blotch and Scratch in Video Based on Video Decomposition

Wang

et al. 2013

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

In old video restoration, automatic detection of common defects, e.g., scratches and blotches, has always been emphasized. While prior thoughts mainly focus on detecting blotches and linear, vertical scratches separately, this paper contributes to a more generalized and challenging issue: simultaneous detection of blotches and complex scratches in video, with much less knowledge of them. We investigate the characteristics of blotches and scratches in space and time domain, and propose a novel detection method based on two main steps: cartoontexture decomposition in the space domain and content-defect separation in the time domain. We then formulate it into convex optimization problems and develop corresponding algorithms. The experiment results demonstrate that the proposed method is of high detection accuracy, verifying the effectiveness of our detection via a video decomposition method.

show abstract

Section: A Prior Workmentioning

confidence: 99%

Section: B Parameter Settingmentioning

confidence: 99%

Detection of Blotch and Scratch in Video Based on Video Decomposition

Wang

et al. 2013

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…In particular, we avoid the difficult task of tracking true scratches, whose temporal behaviour is difficult to determine. In order to decide on the rejection of a detection, we also estimate a robust affine scene motion, in contrast to some previous methods [9], [19] which employ less robust motion estimation.…”

Section: Introductionmentioning

confidence: 99%

Robust Automatic Line Scratch Detection in Films

Newson

Almansa

Gousseau

et al. 2014

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Line scratch detection in old films is a particularly challenging problem due to the variable spatiotemporal characteristics of this defect. Some of the main problems include sensitivity to noise and texture, and false detections due to thin vertical structures belonging to the scene. We propose a robust and automatic algorithm for frame-by-frame line scratch detection in old films, as well as a temporal algorithm for the filtering of false detections. In the frame-by-frame algorithm, we relax some of the hypotheses used in previous algorithms in order to detect a wider variety of scratches. This step's robustness and lack of external parameters is ensured by the combined use of an a contrario methodology and local statistical estimation. In this manner, over-detection in textured or cluttered areas is greatly reduced. The temporal filtering algorithm eliminates false detections due to thin vertical structures by exploiting the coherence of their motion with that of the underlying scene. Experiments demonstrate the ability of the resulting detection procedure to deal with difficult situations, in particular in the presence of noise, texture, and slanted or partial scratches. Comparisons show significant advantages over previous work.

show abstract

“…• Primary scratch with length equal to height of frame or 90% of frame height [2] • Secondary scratch of length less than the height of primary scratch. • The scratch width is usually less than 5 pixels.…”

Section: Introductionmentioning

confidence: 99%

Line scratch restoration using flow vectors of spatio-temporal video cube

Kumar

Gupta

Venkatesh

2014

2014 Annual IEEE India Conference (INDICON)

View full text Add to dashboard Cite

Line scratch is a prominent artefact in old films. In this paper, restoration of line scratch is proposed using spatio -temporal video cube. Video is considered as a cube and we assume that , and are three flow directions of frames , and respectively. Vertical and horizontal line scratch is a line of some width in frame, but it appears as a rectangular patch in or frame. Such type of patches produce a discontinuity in the flow vector of these frames. Restoration method is divided into two steps. In the first step the discontinuity in flow vector is smoothed by applying a moving average filter and corrupted pixels in the patch area are recovered by transforming the backward and forward frames. Most of the missing data is recovered by this approach. A median filter is then applied on the scratch location of frame for smoothing the pixels. The proposed spatio-temporal technique has removed vertical and horizontal line scratches in old film video. The comparison of the proposed algorithm with existing methods validates the improved performance of our result.

show abstract

Scratch detection supported by coherency analysis of motion vector fields

Cited by 13 publications

References 12 publications

Detection of Blotch and Scratch in Video Based on Video Decomposition

Detection of Blotch and Scratch in Video Based on Video Decomposition

Robust Automatic Line Scratch Detection in Films

Line scratch restoration using flow vectors of spatio-temporal video cube

Contact Info

Product

Resources

About