Spatial surface coarseness analysis: technique for fingerprint spoof detection

Pereira, Luis F. Alves; Pinheiro, Hector N. B.; Cavalcanti, George D. C.; Ren, Tsang Ing

doi:10.1049/el.2012.4173

Cited by 18 publications

(13 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the results in Table 3 we choose the parameter pair ( 16,4) to compare with the previous methods.…”

Section: Comparison With Previous Methodsmentioning

confidence: 99%

“…The quality features were extracted by ridge-strength, ridge-clarity and ridge-continuity measures. In 2013, Pereira et al [16] measured the coarseness of fingerprint through the estimation of the residual Gaussian white noise of the image. The noise was divided into several parts, and each part of the noise was used to calculate a histogram.…”

Section: Related Workmentioning

confidence: 99%

“…The detection accuracies of the classifiers are listed in Table 3. The parameter pairs ( , ) used in the training process of classifiers are presented in Table 4 As shown in Table 3, the feature vectors extracted with the parameter pairs (8,5) and (16,4) achieve the best average detection result. When the quantization factor is set to 4, we need to set the truncation factor to 10 so as to cover 91.43% of gradients, generating 882 features.…”

Section: Selection Of Parametersmentioning

confidence: 99%

See 2 more Smart Citations

Spoof Fingerprint Detection based on Co-occurrence Matrix

Jiang¹,

Liu²

2015

IJSIP

View full text Add to dashboard Cite

“…According to the results in Table 3 we choose the parameter pair ( 16,4) to compare with the previous methods.…”

Section: Comparison With Previous Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Selection Of Parametersmentioning

confidence: 99%

See 1 more Smart Citation

Spoof Fingerprint Detection based on Co-occurrence Matrix

Jiang¹,

Liu²

2015

IJSIP

View full text Add to dashboard Cite

“…In biometric recognition, fingerprint recognition has become the most mature and extensive because of the characteristics: uniqueness, stability, easy to acquire, etc. [2].…”

Section: Introductionmentioning

confidence: 99%

Lightweight Convolutional Neural Network Based on Singularity ROI for Fingerprint Classification

Wen

Zhou

2020

IEEE Access

View full text Add to dashboard Cite

Fingerprint classification is a significant guarantee for efficient and accurate fingerprint recognition, especially when dealing with one-to-many fingerprint recognition. However, due to large intra-class variability, small inter-class variability, and noise, existing fingerprint classification algorithms still require further improvement in performance and efficiency. In this paper, a Lightweight CNN (Convolutional Neural Network) structure based on singularity ROI (region of interest) is proposed. The experimental results show that the accuracy on testing set of the proposed structure achieves 93%, which is far better than classic non-NN (neural network) classifiers, including RF(Random Forest), KNN (K-Nearest-Neighbor), LR (Logistic Regression), Linear SVM (Support Vector Machine), and RBF (Radial Basis Function) SVM. More momentously, compared with other three CNN structures published in recent years, the proposed structure achieves similar or even better performance with 1/12 to 1/38 parameter scale of other structures, which helps to proceed faster training and testing. Moreover, the proposed CNN model with fewer neurons can achieve better suppression of overfitting and robustness to noise.

show abstract

“…It was used in 2005 [1] to compute the global coarseness of the image, which was used as a discriminative feature. This approach has been recently improved in [2] by computing coarseness on disjoint blocks and generating a histogram-based feature vector for support vector machine (SVM) classification. This local analysis allows a significant increase in performance, enlightening one to the fact that the features computed on a patch-basis can be more effective in discrimination, especially on the low-resolution images such as those generated by common scanners.…”

mentioning

confidence: 99%