Banca de QUALIFICAÇÃO: GABRIEL LUCAS COSTA PEREIRA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : GABRIEL LUCAS COSTA PEREIRA
DATE: 23/01/2026
TIME: 14:00
LOCAL: https://meet.google.com/kyx-ngad-jht
TITLE:

CGRA ACCELERATOR FOR IMAGE PROCESSING IN NANOSATELLITES

KEY WORDS:

Deep Neural Networks. CGRA. Hardware Accelerator. Reconfigurable

Architectures. Nanosatellites. Image Processing.

PAGES: 63
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Sistemas de Computação
SPECIALTY: Hardware
SUMMARY:

In the current scenario, intelligent systems are being processed at the edge, and architectural

solutions and accelerators are emerging as alternatives to the previously continuously scalable

Moore's Law. These intelligent systems are adopting applications with deep neural networks

(DNNs) for greater efficiency, and they require low power consumption and often real-time

responses. Given the scenario, reconfigurable and coarse-grained architectures such as

CGRAs (Coarse Grained Reconfigurable Arrays) can be considered as excellent solutions for

optimal performance along with lower power and area consumption compared to other

accelerators commonly disseminated in the market such as FPGAs and GPUs. This work

proposes an architectural implementation of a CGRA accelerator for the case study of image

processing in nanosatellites using CNNs and employing the reconfigurability of CGRA for

conventional image processing as well as for neural networks. This implementation is

accompanied by an analysis of hardware architectural optimization techniques and neural

networks found in the literature.

COMMITTEE MEMBERS:
Presidente - 1694485 - MARCIO EDUARDO KREUTZ
Interna - 1882699 - MONICA MAGALHAES PEREIRA
Externo à Instituição - SILVIO ROBERTO FERNANDES DE ARAUJO - UFERSA