::: Anderson Rocha .:. Institute of Computing (UNICAMP), Brazil :::


Ongoing Research Projects
Diabetic Retinopathies \| Forgery Detection
In the following you can find some of the research projects I am involved with (either as an associate or principal investigator).

*Automatic screening of Diabetic Retinopathies: IT in the fight against preventable blindness* Researchers Involved - Jacques Wainer (principal investigator, University of Campinas) - Anderson de Rezende Rocha (University of Campinas) - Siome Klein Goldenstein (University of Campinas) - Ivan Pisa (Federal University of São Paulo) - Michel Eid Farah (Federal University of São Paulo) - Paulo Schor (Federal University of São Paulo) Period: 2009-2010. Abstract The development of an automatic screening program for diabetic retinopathies (DR), and its application on a teleoftalmology service. The system will detect changes in digital photos of the back of the eye which indicate some degree of DR, and it will then forward the patient's images to an specialist for diagnosis. If there are no alterations in the retina image, the patient will not be forwarded to the specialist. At first the system will detect the following alterations: hemorrhages, exsudates, scars, hiper- and hypo-pigmentation. Grant FAPESP - Fundação de Amparo a Pesquisa do Estado de São Paulo Microsoft Inc. Related publications: please access my publications area.
			Up

*Digital Image Forensics: forgery and spoofing detection* Researchers Involved - Anderson de Rezende Rocha (Principal investigator, University of Campinas) - Siome Klein Goldenstein (University of Campinas) Period: 2009. Abstract Image tampering detection is paramount nowadays. Digital image forensics encompasses the set of tools and techniques to verify authenticity and integrity of digital images with several applications in our life. For instance, we can deploy digital forensics solutions to improve the security on biometric systems such as face and fingerprint recognition where an attacker may want to have unauthorized access to the system pretending to be someone else. The financial system is not safe either. Some observers consider check frauds as one of the most increasing crimes of the 1990s. In this context, it is necessary to devise afficient and effective automatic tools to detect such forgeries. Particularly, we are interested in designing solutions for such problems borrowing ideas from Machine Learning, Image Processing, and Computer Vision. We envision solutions related to check fraud detection as well as face and fingerprint spoofing detection. In addition, our work fits the main directions proposed in the thematic project FAPESP 07/52015-0 - Approximation methods for visual computing - and it is a joint initiative with University of Colorado at Colorado Springs, United States. Grant FAPESP - Fundação de Amparo a Pesquisa do Estado de São Paulo Related publications: please access my publications area.
			Up

Ongoing Research Projects