Smart camera | EPFL Graph Search

A smart camera (sensor) or intelligent camera (sensor) or (smart) vision sensor or intelligent vision sensor or smart optical sensor or intelligent optical sensor or smart visual sensor or intelligent visual sensor is a machine vision system which, in addition to image capture circuitry, is capable of extracting application-specific information from the captured images, along with generating event descriptions or making decisions that are used in an intelligent and automated system. A smart camera is a self-contained, standalone vision system with built-in in the housing of an industrial video camera. The vision system and the image sensor can be integrated into one single piece of hardware known as intelligent image sensor or smart image sensor. It contains all necessary communication interfaces, e.g. Ethernet, as well as industry-proof 24V I/O lines for connection to a PLC, actuators, relays or pneumatic valves, and can be either static or mobile. It is not necessarily larger than

Image processing on reconfigurable hardware for continuous monitoring of fluorescent biomarkers in cell cultures

Julien Michel Ghaye

Fluorescence microscopy is a widespread tool in biological research. It is the primary modality for bioimaging and empowers the study and analysis of multitudes of biological processes. It can be applied to fixed biosamples, that is samples with frozen biological features by mean of chemical linkers, or live biosamples providing useful insights on the spatio-temporal behavior of fluorescently stained biomarkers. Current fluorescent microscopy techniques use digital image sensors which are used to leverage quantitative studies instead qualitative outcomes. However, state-of-the-art techniques are not suitable for integration in small, contained and (semi-)autonomous systems. They remain costly, bulky and rather quantitatively inefficient methods for monitoring fluorescent biomarkers, which is not on par with the design constraints found in modern Lab-on-a-Chip or Point-of-Use systems requiring the use of miniaturized and integrated fluroscence microscopy. In this thesis, I summarize my research and engineering efforts in bringing an embedded image processing system capable of monitoring fluorescent biomarkers in cell cultures in a continuous and real-time manner. Three main areas related to the problem at hand were explored in the course of this work: simulation, segmentation algorithms and embedded image processing. n the area of simulation, a novel approach for generating synthetic fluorescent 2D images of cell cultures is presented. This approach is dichotomized in a first part focusing on the modeling and generation of synthetic populations of cells (i.e. cell cultures) at the level of single fluorescent biomarkers and in a second part simulating the imaging process occurring in a traditional digital fluorescent microscope to produce realistic images of the synthetic cell cultures. The objective of the proposed approach aims at providing synthetic data at will in order to test and validate image processing systems and algorithms. Various image segmentation algorithms are considered and compared for the purpose of segmenting fluorescent spots in microscopic images. The study presented in this thesis includes a novel image thresholding technique for spot extraction along with three well-known spot segmentation techniques. The comparison is undertaken on two aspects. The segmentation masks provided by the methods are used to extract further metrics related to the fluorescent signals in order to (i) evaluate how well the segmentation masks can provide data for classifying real fluorescent biological samples from negative control samples and (ii) quantitatively compare the segmentations masks based on simulated data from the previously stated simulation tool. Finally, the design of an embedded image processing system based on FPGA technologies is showcased. A semi-autonomous smart camera is conceived for the continuous monitoring of fluorescent biomarkers based on one of the segmentation methods incorporated in the previously stated comparison. Keeping the focus on the need for integration in fluorescence microscopy, the image processing core at the heart of the smart camera results from the use of a novel image processing suite; a suite of IP cores developed under the constraints dictated by the bioimaging needs of fluorescence microscopy for use in FPGA and SoC technologies. As a proof of concept, the smart camera is applied to the monitoring of the kinetics of the uptake of fluorescent silica nano-particles in cell cultures.

EPFL2015

A Real-time Multi-aperture Omnidirectional Visual Sensor Based on Interconnected Network of Smart Cameras

Hossein Afshari, Abdulkadir Akin, Ömer Cogal, Yusuf Leblebici, Vladan Popovic, Alexandre Schmid, Kerem Seyid

Centralized and multi-level implementations of the Panoptic omnidirectional multiaperture visual system were previously presented by the authors, relying on transmission of all camera outputs to a single central processing node for omnidirectional image and video reconstruction. In this paper, a novel distributed and parallel implementation of the omnidirectional vision reconstruction algorithm of the Panoptic system is presented. The parallel approach aims to overcome the scalability problems and memory bandwidth limitations of the centralised approach. The real-time hardware implementation is presented for camera modules with image processing, memory and interconnectivity features. A methodology is introduced for the arrangement of camera modules with interconnectivity feature into a target interconnection network topology. A unique custom-made multiple- FPGA hardware platform is introduced for the implementation of an interconnected network of 49 camera prototype Panoptic system. A hardware architecture based on presented hardware platform enabling the real-time implementation of the blending algorithms is presented, along with the imaging results and resource utilization. The real-time implementation results of the implemented omnivision application on the aforementioned prototype are demonstrated.

Institute of Electrical and Electronics Engineers2015

Video surveillance using JPEG 2000

Frédéric Dufaux, Touradj Ebrahimi

This paper describes a video surveillance system which is composed of three key components, smart cameras, a server, and clients, connected through IP-networks in wired or wireless configurations. The system has been designed so as to protect the privacy of people under surveillance. Smart cameras are based on JPEG 2000 compression where an analysis module allows for events detection and regions of interest identification. The resulting regions of interest can then be encoded with better quality and scrambled. Compressed video streams are scrambled and signed for the purpose of privacy and data integrity verification using JPSEC compliant methods. The same bitstream may also be protected for robustness to transmission errors based on JPWL compliant methods. The server receives, stores, manages and transmits the video sequences on wired and wireless channels to a variety of clients and users with different device capabilities, channel characteristics and preferences. Use of seamless scalable coding of video sequences prevents any need for transcoding operations at any point in the system.

2004

Image processing on reconfigurable hardware for continuous monitoring of fluorescent biomarkers in cell cultures

Julien Michel Ghaye

EPFL2015