01-03-2011, 09:18 AM
presented by:
K.Muralidhar
P.Sindhuri
@@@@@.pptx (Size: 1.59 MB / Downloads: 104)
Smart cameras in embedded systems
ABSTRACT
Smart cameras are rapidly finding their way into intelligent surveillance systems
Recognizing faces in the crowd in real-time is one of the key features that will significantly enhance intelligent surveillance systems
The main challenge is the fact that the enormous volumes of data generated by high-resolution sensors can make it computationally impossible to process on mainstream processors
INTRODUCTION
Video surveillance is becoming more and more essential now-a-days as society relies on video surveillance to improve security and safety
For security, such systems are usually installed in areas where crime can occur such as banks and car parks.
For safety, the systems are installed in areas where there is the possibility of accidents such as on roads or motorways and at construction sites
Difference between digital cameras and smart cameras
Digital cameras just capture images
Smart cameras capture high level descriptions of the scene and analyze what they see
Improving smart camera design
High resolution image sensor
High bandwidth communication interface
Reconfigurable platform for hardware and software processors
Improved Smart Camera Design architecture
High Resolution Image Sensor
Overall Scene (a), ROI extracted from scene with resolution of 7MP (b), 5 MP©, 3MP (d), 1MP (e) and VGA (f).
Detection and Recognition Algorithms
Our algorithms use both-
A) Low-level processing
B) High-level processing
A) Low-level processing
Region extraction
Contour following
Ellipse fitting
Graph matching
B) High-level processing
The high-level processing component, which can be adapted to different applications, compares the motion pattern of each body part—described as a spatiotemporal sequence of feature vectors—in a set of frames to the patterns of known postures and gestures and then uses several hidden Markov models in parallel to evaluate the body's overall activity.
Requirements
Frame rate
“the system must process a certain amount of frames per second to properly analyze motion and provide useful results”
Latency
“the amount of time it takes to produce a result for a frame is also important because smart cameras will likely be used in closed loop control systems where high latency makes it difficult to initiate events in a timely fashion based on action in the video field”
K.Muralidhar
P.Sindhuri
@@@@@.pptx (Size: 1.59 MB / Downloads: 104)
Smart cameras in embedded systems
ABSTRACT
Smart cameras are rapidly finding their way into intelligent surveillance systems
Recognizing faces in the crowd in real-time is one of the key features that will significantly enhance intelligent surveillance systems
The main challenge is the fact that the enormous volumes of data generated by high-resolution sensors can make it computationally impossible to process on mainstream processors
INTRODUCTION
Video surveillance is becoming more and more essential now-a-days as society relies on video surveillance to improve security and safety
For security, such systems are usually installed in areas where crime can occur such as banks and car parks.
For safety, the systems are installed in areas where there is the possibility of accidents such as on roads or motorways and at construction sites
Difference between digital cameras and smart cameras
Digital cameras just capture images
Smart cameras capture high level descriptions of the scene and analyze what they see
Improving smart camera design
High resolution image sensor
High bandwidth communication interface
Reconfigurable platform for hardware and software processors
Improved Smart Camera Design architecture
High Resolution Image Sensor
Overall Scene (a), ROI extracted from scene with resolution of 7MP (b), 5 MP©, 3MP (d), 1MP (e) and VGA (f).
Detection and Recognition Algorithms
Our algorithms use both-
A) Low-level processing
B) High-level processing
A) Low-level processing
Region extraction
Contour following
Ellipse fitting
Graph matching
B) High-level processing
The high-level processing component, which can be adapted to different applications, compares the motion pattern of each body part—described as a spatiotemporal sequence of feature vectors—in a set of frames to the patterns of known postures and gestures and then uses several hidden Markov models in parallel to evaluate the body's overall activity.
Requirements
Frame rate
“the system must process a certain amount of frames per second to properly analyze motion and provide useful results”
Latency
“the amount of time it takes to produce a result for a frame is also important because smart cameras will likely be used in closed loop control systems where high latency makes it difficult to initiate events in a timely fashion based on action in the video field”