A Caltech Library Service

Non-linear matched filtering for point source detection

Makovoz, David (2005) Non-linear matched filtering for point source detection. In: Image Processing: Algorithms and Systems IV. Proceedings of SPIE. No.5672. Society of Photo-optical Instrumentation Engineers (SPIE) , Bellingham, WA, pp. 358-369. ISBN 9780819456458.

[img] PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


The task of object detection depends on the ability to suppress the noise present in images in order to increase the signal-to-noise ratio. The standard linear matched filter is the optimal filter on the assumption of the Gaussian distribution of the signal and the noise. However, as a rule the distribution of the signal in image processing is not Gaussian. The linear matched filter becomes sub-optimal. Any non-Gaussian distribution function can be closely approximated using the Gaussian Mixture Model (GMM). We use GMM to approximate the signal distribution function and derive the optimal filter by means of mean square error (MSE) minimization. The optimal non-linear filter is determined by the assumed signal distribution function. We use non-linear matched filtering for point source detection in astronomical images. We derive the GMM components by fitting the theoretical point source distribution function. The filtered images are subjected to image segmentation and subsequent point source detection. The non-linear matched filtering has been tested with simulated data and has been shown to significantly improve the quality of point source detection. Receiver operating characteristic technique has been used to evaluate performance of various Gaussian mixtures for point source detection. This algorithm is currently used for the Spitzer Spatial Telescope.

Item Type:Book Section
Related URLs:
URLURL TypeDescription
Alternate Title:Nonlinear matched filtering for point source detection
Additional Information:© 2005 Society of Photo-Optical Instrumentation Engineers (SPIE). I would like to thank the MIPS and IRAC Instrument Support Teams in general, and David Frayer in particular, for reducing the data and making the MIPS 70 micron PRF. I also would like to thank David Shupe for creating the simulated images used in this work. This work was carried out at the Spitzer Science Center, with funding from NASA under contract 1407 to the California Institute of Technology and the Jet Propulsion Laboratory.
Group:Infrared Processing and Analysis Center (IPAC)
Funding AgencyGrant Number
Subject Keywords:infrared imaging, matched filtering, non-linear filtering, non-Gaussian distributions, non-linear matched filtering, object detection, Gaussian mixture model, receiver operating characteristic
Series Name:Proceedings of SPIE
Issue or Number:5672
Record Number:CaltechAUTHORS:20190221-110528585
Persistent URL:
Official Citation:David Makovoz "Nonlinear matched filtering for point source detection", Proc. SPIE 5672, Image Processing: Algorithms and Systems IV, (1 March 2005); doi: 10.1117/12.587283;
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:93153
Deposited By: George Porter
Deposited On:22 Feb 2019 15:32
Last Modified:16 Nov 2021 16:55

Repository Staff Only: item control page