Feature extraction of non-stochastic surfaces using curvelets

Li, Linfu; Zhang, Xiangchao; Zhang, Hao; He, Xiaoying; Xu, Min

doi:10.1016/j.precisioneng.2014.09.003

Cited by 10 publications

(4 citation statements)

References 19 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It provides an efficient representation of smooth objects with discontinuities along curves. As an extension of the wavelet concept, the curvelet transform was not only successfully applied in several image processing fields but also outperformed wavelet-based methods [19,20]. To perform fast discrete curvelet transform, two implementations were proposed [21].…”

Section: Curvelet Transformmentioning

confidence: 99%

Breast cancer diagnosis in digitized mammograms using curvelet moments

Dhahbi

Barhoumi

Zagrouba

2015

Computers in Biology and Medicine

107

View full text Add to dashboard Cite

Section: Curvelet Transformmentioning

confidence: 99%

Breast cancer diagnosis in digitized mammograms using curvelet moments

Dhahbi

Barhoumi

Zagrouba

2015

Computers in Biology and Medicine

107

View full text Add to dashboard Cite

“…Zhang, Li, and Li [19] proposed a curvelet transform based method to improve Canny edge operator for edge detection in tire laser shearography images. Some studies proposed mixed methods based on curvelet transform for classifications, such as combining with morphological feature extraction method to characterize non-stochastic surfaces for metrology [20], fusing with Kirsch's templates to extract retinal blood vessels for detection of diabetes at early stages [21], integrating with PCA method to extract features from still images for face recognition [22].…”

Section: Literature Reviewmentioning

confidence: 99%

“…These subimages greatly reduces the dimensionality of the original image. Thereafter, only the approximate components are selected to perform further computations, as they account for maximum variance [20,29]. Thus, a representative and effi cient feature set is produced.…”

Section: Feature Extractionmentioning

confidence: 99%

Creation of image models for inspecting defects in commercial dried fish floss

Lin¹,

Lin²,

Lin³

2020

J Appl Eng Science

View full text Add to dashboard Cite

Fish fl oss is a chopped fi nely or mashed fi sh meat boiled in seasonings, then stir fry until the fi sh meat is arid and pulverous. In the making of commercial fi sh fl osses, defect inspection is conducted by expertise inspectors using their feeling of contact and sight which could cause misjudgments. When consumers eat fi sh fl oss with defects, it may cause harm to the health of consumers. Therefore, this study proposes an automated defect detection method and develop an optical inspection system for commercial dried fi sh fl oss. The proposed method applies the curvelet transform with low-pass energy fi ltering to remove the random patterns of background and delete the angle direction of background texture. The approximated and partial detailed components regarding defects and uniform background are preserved in the low and medium frequency bands. In the reconstructed image, the background random texture is attenuated and the defect areas are enhanced. Finally, the restored image can be easily segmented by an estimated threshold value into two categories namely dark defects, and white background. The experimental results show that the proposed method well balances the trade-off between the recall rate (82.11%) and precision rate (87.62%), and reaches an F-score of 84.78%, outperforming the traditional defect detection techniques in inspection of dried fi sh fl oss.

show abstract

“…Extraction and subsequent assessments of selected surface topography features can be perceived as being of remarkable value when tribological behavior of machined parts is taken into consideration in more detail. Moreover, feature-based characterization of surface textures was performed for many tribological performances [1,2].…”

Section: Introductionmentioning

confidence: 99%

Metrology and Measurement Systems

Podulka

2020

View full text Add to dashboard Cite

In this paper both envelope approach and morphological filters for characterisation of surface textures were proposed, applied and thoroughly examined. Obtained results were compared with those received after appliance of commonly-used algorithms. The effect of appliance of proposed procedures on surface topography parameters (from ISO 25178 standard) was taken into consideration. The following types of surface textures were assessed: two-process plateau-honed cylinder liners, plateau-honed cylinder liners with additionally burnished dimples, turned piston skirts, grinded and/or isotropic topographies. It was assumed that envelope characteristics (envelope filtration) can provide results useful for assessments of deep and/or wide oil-reservoirs especially when they are edge located. Moreover, some near-valley areas of surface texture details can be less distorted when envelope filtering is accomplished. It was also found that closing and/or opening envelope filtration can be valuable for reduction of some surface topography measurement errors.

show abstract

Feature extraction of non-stochastic surfaces using curvelets

Cited by 10 publications

References 19 publications

Breast cancer diagnosis in digitized mammograms using curvelet moments

Breast cancer diagnosis in digitized mammograms using curvelet moments

Creation of image models for inspecting defects in commercial dried fish floss

Metrology and Measurement Systems

Contact Info

Product

Resources

About