Client threshold prediction in biometric signature recognition by means of Multiple Linear Regression and its use for score normalization

Vivaracho-Pascual, Carlos; Simon-Hurtado, Arancha; Manso-Martinez, Esperanza; Pascual-Gaspar, J. M.

doi:10.1016/j.patcog.2016.02.007

Cited by 7 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results show that the score ratio proposal can be applied with small cohort sets and with small N values, which is very important for real applications. 2 The percentages have been calculated as follow:…”

Section: Results Analysismentioning

confidence: 99%

See 1 more Smart Citation

Using the score ratio with distance-based classifiers: A theoretical and practical study in biometric signature recognition

2017

Self Cite

View full text Add to dashboard Cite

Section: Results Analysismentioning

confidence: 99%

“…This work is focused on the Match stage. We have approached the Feature Extraction and Decision stages in previous works, [1] and [2] .…”

Section: Introductionmentioning

confidence: 99%

Using the score ratio with distance-based classifiers: A theoretical and practical study in biometric signature recognition

2017

Self Cite

View full text Add to dashboard Cite

“…As a result of this observation, studies on the quality of biometric samples emerged in order to improve system performance and reliability. For instance, based on a quality measure, one could decide whether to: 1) alter the enrollment process including requesting users to re-enroll in the system, 2) adjust the decision threshold [39] or weighting function in a multimodal biometric system, 3) invoke different preprocessing or recognition algorithms [2], or 4) update the template [8]. A quality score could also be used as a deciding factor whether an additional biometric trait has to be required [28].…”

Section: Related Workmentioning

confidence: 99%

“…As a result, attackers could easily produce a signature that matches the template, resulting in a compromised account. In this context, an assessment of biometric template characteristics could be used to design proper mechanisms to cope with such weak templates [2,13,28,39]. For example, given a template that is predicted to yield high FAR (False Acceptance Rate: the likelihood that forgery samples will be incorrectly accepted by the system) but low FRR (False Rejection Rate: the likelihood that genuine samples will be incorrectly rejected by the system), the system could examine whether a few bad enrolled samples could be safely removed to lower FAR without degrading FRR.…”

Section: Introductionmentioning

confidence: 99%

Distinctiveness, complexity, and repeatability of online signature templates

Sae-Bae¹,

Memon²,

Sooraksa³

2018

Pattern Recognition

View full text Add to dashboard Cite

This paper proposes three measures to quantify the characteristics of online signature templates in terms of distinctiveness, complexity and repeatability. A distinctiveness measure of a signature template is computed from a set of enrolled signature samples and a statistical assumption about random signatures. Secondly, a complexity measure of the template is derived from a set of enrolled signature samples. Finally, given a signature template, a measure to quantify the repeatability of the online signature is derived from a validation set of samples. These three measures can then be used as an indicator for the performance of the system in rejecting random forgery samples and skilled forgery samples and the performance of users in providing accepted genuine samples, respectively. The effectiveness of these three measures and their applications are demonstrated through experiments performed on three online signature datasets and one keystroke dynamics dataset using different verification algorithms.

show abstract

“…The ensemble group did not demand normalization of the dataset, once that normalization is a preprocessing step that uses tree-based approach, like RF and ADAB, do not require, even so, other MLs based on matrix distance as the regularizations and MLR are necessary(VIVARACHO-PASCUAL et al 2016). …”

mentioning

confidence: 99%

Body weight prediction of crossbred beef cattle through the image processing and machine learning algorithms

Silva¹

View full text Add to dashboard Cite

For the beef cattle system, one of the most valuable information is the body weight that can be linked to animal growth and performance. The bidimensional sensors area is the cheapest technology among all sensors used as a tool to extract information that can be applied to machine learning to predicts value phenotype. This study aimed to predict body weight using video image analysis with simple bidimensional equipment, from the dorsal view of crossbreed beef cattle (1⁄2 Angus x 1⁄2 Nellore) in a finishing system, applying different frame information and machine learning algorithms. The experimental procedures were performed at Federal University of Viçosa. A total of 40 crossbreed steers (1⁄2 Angus x 1⁄2 Nellore) were used, averaging 8 months of age at the beginning of the feedlot trial, and 291.7±23.8 kg and 517.42±54.8kg of initial and final body weight, respectively. The data collection occurred from September (12 Month) to December/2021 (15 month). Body weight (BW) was collected using an automatic Intergado company drink fountain/scale and the video images were collected using cameras Intelbras from the animals’ dorsal view. Three approaches were tried for segmentation of the animals’ dorsal images, however, their color characteristics did not allow do this automatically, so were used ImageJ software to manually do the delimitations, extracting 8 Shape Descriptors. For regression were used 6 machine learning algorithms, Ridge, Least Absolute Shrinkage and Selection Operator (LASSO), Elastic Net (ENET), Multiple Linear Regression (MLR), Adaboost (ADAB) and Random Forest (RF) to construct predictive models, the dataset was split in 70:30 for training and test. The regularizations RIDGE and MLR without AGE as a predictor had similar performance. The AGE addition improved all algorithms, the best metrics results were for ENET and ENET using AGE for a dataset with 5 Frames information (5F) R2=0.68 and 0.76, respectively. Thus, the use of bidimensional sensor in the dorsal view can predict the BW of crossbreed (1⁄2 Angus x 1⁄2 Nellore). Keywords: Correlation. Image Processing Regularization

show abstract

Client threshold prediction in biometric signature recognition by means of Multiple Linear Regression and its use for score normalization

Cited by 7 publications

References 29 publications

Using the score ratio with distance-based classifiers: A theoretical and practical study in biometric signature recognition

Using the score ratio with distance-based classifiers: A theoretical and practical study in biometric signature recognition

Distinctiveness, complexity, and repeatability of online signature templates

Body weight prediction of crossbred beef cattle through the image processing and machine learning algorithms

Contact Info

Product

Resources

About