No abstract

We present in this paper a new system based on Motion JPEG2000 intended for road surveillance application. The system uses a reference image and consists in 4 processing steps, namely initialization phase where the first reference image is built, reference estimation, motion segmentation (foreground extraction, ROI mask), and JPEG2000 coding. A first order recursive filter is used to build a reference image that corresponds to the background image. The obtained background is sent to the decoder once for all. The reference image at the coder side is estimated according to a Gaussian mixture model. The remote reference image is updated when specific conditions are met. The updating remote reference is triggered according to the states of mobile objects in the scene (no, few or lot of mobiles). The motion detection given by classical background subtraction technique is performed in order to extract a binary mask. The motion mask gives the region of interest of the system. The JPEG2000 image coded with a ROI option is sent towards the decoder. The decoder receives, decodes the image and builds the implicit binary ROI mask. Then, the decoder builds the displayed image using the reference image, the current image and the mask.

This paper deals with a new codec based on the JPEG 2000 standard that will use a market hardware codec in order to build a road surveillance device. The developed coder consists in 4 processing steps, namely construction of a reference image, foreground extraction (ROI mask), encoding with JPEG 2000 and transmission through a wireless device. A …rst order recursive …lter is used to build a reference image that corresponds to the background image and the updated reference image is computed according to a mixture of Gaussians model. The system builds a reference image and transmits it towards a decoder through the GSM network. After the initialization phase, the reference image is updated automatically according to a Gaussian mixture model, and when the ROI can be considered as null, a piece of the updated background image is sent. We perform motion detection in order to extract a binary mask. The motion mask gives the region of interest for the system. The current image and the motion mask are coded using the ROI option of JPEG 2000 codec with a very low bit rate and transmitted towards the decoder. The complete scheme is implemented and it reaches the expected performances. We also showed how the local background image is built and updated at each frame. We presented the strategy in order to update smoothly the remote background image. The implementation runs at 5-8 frames per second on a 1.8 GHz AMD processor for 320x240 color images. Telephone: +33 (0)5 59 43 85 05Magys is a SME involved in developing video devices mainly for video surveillance.

No abstract

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).