Editing Image Analysis in Astronomy for Very Large Vision Machine

Title: Image Analysis in Astronomy for Very Large Vision Machine

Authors: Gerardo Iovane

Category: R&D/Lab Automation

Products Used:
- LABVIEW 6i Prof Dev Sys ver 6.02
- IMAQ Vision 6.0
- SQL TOOLKIT
- INTERNET TOOLKIT
- SIGNAL PROCESSING TOOLSET

Challenge: Develop a very fast and real-time system to perform image acquisition, reduction, and analysis for detecting very faint luminosity variations, related to the discovery of new planets outside the Solar System.

Solution: A real-time network parallel image processing system is developed using Labview. The architecture is provided with a server, linked with a maximum of 25 workstations to analyze even 256 Mpixels images. The server takes data directly from the acquisition system, composed of a telescope with 16k × 16k pixels CCD camera.

Abstract: A very complex system (hardware/software) is developed to detect luminosity variations connected with the discovery of new planets outside the Solar System. Traditional imaging approaches are very demanding in terms of computing time; therefore, an automatic vision and decision software architecture is presented. It allows performing an online discrimination of interesting events by using two levels of triggers. A fundamental challenge was to work with very large CCD cameras (even 16k × 16k pixels) in line with very large telescopes. The architecture can use a distributed parallel network system based on a maximum of 256 standard workstations.

Introduction: During the last ten years, much attention has been devoted to planet detection. The passage of a planet close to the line of sight of the observer implies a luminosity variation of the star. The presented architecture was implemented to give an answer to this challenge. There are many difficulties due to the large amount of monitored luminous objects, corresponding to never realized CCD cameras (even 256 Mpixels per image). To get an effective solution, it was designed a parallel system, distributed in the world, that can analyze data and pre-processed images in real-time.

Environment: The system architecture is structured as shown in Fig. 1, while the Server interface is presented in Fig. 2. All software is realized with Labview. In addition, image analysis is realized with IMAQ and data analysis uses the Signal Processing Toolset by NI.

The server machine controls the following units:

1. Image Data Acquisition (I-DAQ) Unit: responsible for data acquisition and pre-reduction of data coming from images.
2. Control Unit: thanks to Telescope Control System (TCS), controls the telescope by following the instructions provided by the DataBase Control System (DBCS) or by the user.
3. DataBase (DB) Unit: performs the data storage and processing according to simulations or previous observations. The DB uses SQL language and is linked to the system by SQL TOOLKIT by NI.

The clients contain the following units:

1. Processing and Analyzing (P&A) Unit: the platform where massive data analysis is performed. It consists of three main units: Data Pre-Processing Unit (DAP) for astrometric alignment, photometric calibration, and PSF correction; Data Processing Unit (DP) for peak detection of relevant luminosity variation, and for removing useless objects; Data Analysis Unit (DAU) for the fits of light curve with different expected models, color correlation, χ2 test. Step 7 of the previous paragraph is the most relevant component of the DAP Unit, and it is composed by four sections: a) the peak detection procedure, b) the star detection and filtering algorithm, c) the cosmic rays filter, and d) the peak classification (single, double, multiple peak curve) methods. It is relevant to stress that for these steps, it can be used or the standard automatic procedures (that uses the peak detection method by NI) or the automatic unsupervised learning methods.

Results: The system successfully detects luminosity variations related to the discovery of new planets outside the Solar System. The distributed parallel network architecture allows real-time analysis and reduces computing time, making the system scalable and adaptable to future improvements in technology and data acquisition methods.

Implications: The presented solution has significant implications for the field of astronomy. It provides a practical and efficient method for detecting new planets outside the Solar System, which can lead to groundbreaking discoveries and advancements in our understanding of the universe. Additionally, the system's architecture and scalability make it a promising platform for future research and applications in the field of astronomical imaging and data analysis.

Link to Article: https://arxiv.org/abs/0308038v1
Authors: 
arXiv ID: 0308038v1

[[Category:Computer Science]]
[[Category:System]]
[[Category:Data]]
[[Category:By]]
[[Category:Analysis]]
[[Category:It]]