Michel Cattoen

A New Linear Calibration Method for Paracatadioptric Cameras

We propose a new calibration method for the paracatadioptric cameras using one image of at least three observed lines. This method is based on the Geyer & Daniilidis (2002) one but has two main advantages. First, a geometric distance is used to compute the camera parameters instead of an algebraic distance. Second, it allows to deal with lines that are projected to straight lines or to circular arcs in an unified manner. We provide a geometric interpretation of the algorithm: The line images are firstly projected to a virtual paraboloid, then planes are fitted on these projections and their intersection finally provides the camera parameters. Thanks to this new formulation, the method is also able to deal very efficiently with outliers. We compare results with existing methods from Geyer & Daniilidis and from Barreto & Araujo (2003)

Automatic inspection of simply patterned material in the textile industry

Automatic inspection allows the continuity of the quality of textiles manufacturing, and the optimization of the pinking according to the situation of the defects. A method for defect detection based on a periodicity breaking of the material densitometrical profile is proposed. After the explanation of the method, some important points like texture suppression, image block size choice, and defect characterization are detailed. Finally, the system configuration and the results obtained from the detection stage are introduced.

Automatic system for monitoring fish passage at dams

Devices called fishways or fish passes are constructed in rivers to help migratory fish get over obstacles (dams). There counting windows are used to monitor fish passage by video-based counting. Our goal is to design and construct a vision system to automate this process. Images are taken by a video camera fitted with an electronic shutter in a backlit fishway. They are stored on optical disks in real time but are processed in delayed time. Faced with high volumes of data, a compression is necessary and an electronic board has been designed to accomplish it in real time. The coding method used is based on a run description of binarized images. Then, a tracking process is implemented on a micro-computer to count the fish crossing the pass. It includes fish recognition, which is based on a Bayesian classification process. In order to reduce processing times, recognition operations (labelling, parameter extraction) are accomplished on coded images. Classification results are satisfactory and are improved by the temporal redundancy generated by the tracking process. Image processing time permits the user, on average, to process images faster than they have been stored. Thus there is no data accumulation. At the end of the processing it is possible to edit a result file, to choose a fish, view its crossing images and change its species if wrong.

A Fast Detector of Line Images Acquired by an Uncalibrated Paracatadioptric Camera

We propose a new method to simultaneously detect the images of lines acquired by an uncalibrated paracatadioptric camera and estimate its parameters. This method is very efficient thanks to our new linear formulation of the constraint for the paracatadioptric line images and to some proposed algorithmic improvements. The line images that are straight lines and circular arcs are detected similarly, after being projected to a virtual paraboloid. Robust estimation methods are then used to find which detected line images are consistent with the best set of camera parameters. We provide experimental results that demonstrate the efficiency and robustness of the proposed method

Compact System for Photometric Characterization of Automotive Headlamps

We present a system allowing the determination of the distribution of light at any given distance (e.g. 25m) from a light source, by measuring the intensity and direction of the illumination distribution at a closer distance (e.g. near 20 cm). The measurement principle is based on the acquisition of images taken by a CCD matrix-sensor through an optical system focused at infinity. A sequence of images is obtained by moving the camera at different positions in front of the source. The process involves the accumulation of shifted images and corrections for optics (geometrical distortion and vignetting) and electronic shutter (exposure time). We describe the developed experimental set-up and the calibration procedures needed to obtain absolute photometric values. We present results obtained for automotive headlamps: computed relative illumination distribution is very similar to that obtained using a reference given by a measurement tool using a lightmeter. Absolute calibration of the system is performed from the set of measured points

Compact low-cost unit for photometric testing of automotive headlamps

A novel proposal for low-cost photometric testing of automotive headlamps is presented, using a combination of well-known optical metrology principles. Deflectometric measurement techniques and CCD-based photometry permit the precise measurement of the energy distribution and the direction of propagation on a plane close to the headlamp, while simple image-processing techniques permit the far-field photometric distribution on a distant plane to be reconstructed from the near-field data measured. The setup constructed and the main calibrations involved are presented, and results for different headlamp models are shown. The technique may be applied to light sources other than headlamps in order to calculate far-field distributions on any predefined surface of interest from near-field measurements.

NICOLAU: compact unit for photometric characterization of automotive lighting from near-field measurements

The present works depicts a measurement technique intended to enhance the characterization procedures of the photometric emissions of automotive headlamps, with potential applications to any light source emission, either automotive or non-automotive. A CCD array with a precisely characterized optical system is used for sampling the luminance field of the headlamp just a few centimetres in front of it, by combining deflectometric techniques (yielding the direction of the light beams) and photometric techniques (yielding the energy travelling in each direction). The CCD array scans the measurement plane using a self-developed mechanical unit and electronics, and then image-processing techniques are used for obtaining the photometric behaviour of the headlamp in any given plane, in particular in the plane and positions required by current normative, but also on the road, on traffic signs, etc. An overview of the construction of the system, of the considered principle of measurement, and of the main calibrations performed on the unit is presented. First results concerning relative measurements are presented compared both to reference data from a photometric tunnel and from a plane placed 5m away from the source. Preliminary results for the absolute photometric calibration of the system are also presented for different illumination beams of different headlamps (driving and passing beam).

ORIENT-CAM, a camera that knows its orientation and some applications

We introduce a new type of smart cameras. These cameras have an embedded orientation sensor which provides an estimate of the orientation of the camera. In this paper, we describe our prototype orientation sensor and propose some methods for the calibration of the whole camera. We then show two applications. First, the camera is used to create oriented spherical panoramas. Second, it is used for image based localization, in which only the position of the camera has to be retrieved.