The objective of this paper is to present the results of a class developed with routines in Java language, and contribution of OpenCV library, for analysis and extraction of metadata from images. To evaluate the developed class, three different figures were produced in cardstock and their perimeters were measured with a millimeter ruler. Then these figures were scanned for further image analysis with aid of the developed class. The images of the figures were initially saved in BMP format. After it, each of the images in BMP format were saved in JPG and PNG file formats resulting, at the end, on nine images. The validation of the correct extraction of the image metadata and so the perimeter value of the object was performed by comparing the values obtained by direct measurement perimeter of the figure, with a millimeter ruler, and the values obtained with digital image processing, counting the contour pixels of the image of the figure, and using the image resolution, one of the extracted metadata. For the edge detection and counting of the contour pixels of object, the algorithms cvFindContours() and cvContourPerimeter(), of OpenCV library, were used. It was obtained, for the worst case, a percentage error of 8.0 %, for images with BMP and PNG format. Therefore, the developed class presents satisfactory results and is recommended to extract and calculate measures of an object present in the image.
Usually, in a real-world scenario, few signature samples are available to train an automatic signature verification system (ASVS). However, such systems do indeed need a lot of signatures to achieve an acceptable performance. Neuromotor signature duplication methods and feature space augmentation methods may be used to meet the need for an increase in the number of samples. Such techniques manually or empirically define a set of parameters to introduce a degree of writer variability. Therefore, in the present study, a method to automatically model the most common writer variability traits is proposed. The method is used to generate offline signatures in the image and the feature space and train an ASVS. We also introduce an alternative approach to evaluate the quality of samples considering their feature vectors. We evaluated the performance of an ASVS with the generated samples using three well-known offline signature datasets: GPDS, MCYT-75, and CEDAR. In GPDS-300, when the SVM classifier was trained using one genuine signature per writer and the duplicates generated in the image space, the Equal Error Rate (EER) decreased from 5.71% to 1.08%. Under the same conditions, the EER decreased to 1.04% using the feature space augmentation technique. We also verified that the model that generates duplicates in the image space reproduces the most common writer variability traits in the three different datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.