Impact of melanocytic lesion image databases on the pre-training of segmentation tasks using the UNET architecture

Author(s):  
Laura Raquel Bareiro Paniagua ◽  
Jose Luis Vazquez Noguera ◽  
Luis Salgueiro Romero ◽  
Deysi Natalia Leguizamon Correa ◽  
Diego P. Pinto-Roa ◽  
...  
2018 ◽  
Vol 7 (4) ◽  
pp. 45-55 ◽  
Author(s):  
Nikhil Kumar ◽  
Sunny Behal

Face recognition is considered as one of toughest and most crucial leading domains of digital image processing. The human brain also uses a similar kind of technique for face recognition. When scrutinizing a face, the human brain signifies the result. Aside from AN automatic processing system, this technique is very sophisticated, owing to the image variations on account of the picture varieties in as far as area, size, articulation, and stance. In this article, the authors have used the options of native binary pattern and uniform native binary pattern for face recognition. They compute a number of classifiers on publicly available benchmarked ORL image databases to validate the proposed approach. The results clearly show that the proposed LBP-piece shrewd strategy has outperformed the traditional LBP system.


PLoS ONE ◽  
2018 ◽  
Vol 13 (9) ◽  
pp. e0203459 ◽  
Author(s):  
Stephen J. Gilmore

2017 ◽  
pp. 271-293
Author(s):  
Stavros Alifragkis ◽  
Giorgos Papakonstantinou

2012 ◽  
Vol 7 (2) ◽  
pp. 20 ◽  
Author(s):  
Mary Kandiuk ◽  
Aaron Lupton

Objective – This study assessed the needs for digital image delivery to faculty members in Fine Arts at York University in order to ensure that future decisions regarding the provision of digital images offered through commercial vendors and licensed by the Libraries meet the needs of teaching faculty. Methods – The study was comprised of four parts. A Web survey was distributed to 62 full-time faculty members in the Faculty of Fine Arts in February of 2011. A total of 25 responses were received. Follow-up interviews were conducted with nine faculty members. Usage statistics were examined for licensed library image databases. A request was posted on the electronic mail lists of the Art Libraries Society of North America (ARLIS-L) and the Art Libraries Society of North America Canada Chapter (CARLIS-L) in April 2011 requesting feedback regarding the use of licensed image databases. There were 25 responses received. Results – Licensed image databases receive low use and pose pedagogical and technological challenges for the majority of the faculty members in Fine Arts that we surveyed. Relevant content is the overriding priority, followed by expediency and convenience, which take precedence over copyright and cleared permissions, resulting in a heavy reliance on Google Images Search. Conclusions – The needs of faculty members in Fine Arts who use digital images in their teaching at York University are not being met. The greatest shortcomings of licensed image databases provided by the Libraries are the content and technical challenges, which impede the ability of faculty to fully exploit them. Issues that need to be resolved include the lack of contemporary and Canadian content, training and support, and organizational responsibility for the provision of digital images and support for the use of digital images.


Sign in / Sign up

Export Citation Format

Share Document