Isabelle Quidu

Mine classification using a hybrid set of descriptors

This paper is concerned with the problem of recognition of objects laying on the sea-bed. A high resolution sonar provides high-quality acoustic images of the sea-bed, allowing the classification of objects from their cast shadow. After the segmentation step, a set of features is extracted from the shadow. We propose an approach based on a hybrid set of descriptors, combining features of different origins. We first compute topological parameters: the extent and the elongation. In addition to these classical features, affine moment invariants seem suitable for sonar images. Indeed, under weak perspective conditions, the perspective transformation is well approximated by an affine transformation. A four-dimensional vector is then computed characterizing the shadow. The method has been tested on simulated sonar images.

Change detection using Synthetic Aperture Sonar: Preliminary results from the Larvik trial

In April of 2011, FFI led a sea trial near Larvik, Norway on FFIs research vessel the H.U. Sverdrup II with participation by representatives from Canada, United States, and France. One objective of the sea trial was to acquire a data set suitable for examining incoherent and coherent change detection and automated target recognition (ATR) algorithms applied to Synthetic Aperture Sonar (SAS) imagery. The end goal is to produce an automated tool for detecting recently placed objects on the seafloor. To test these algorithms two areas were chosen, one with a comparatively benign seafloor and one with a boulder strewn complex seafloor. Each area was surveyed before and after deployment of objects. The survey time intervals varied from two days to eight days. In this paper we present the trial and show examples of SAS images and change detection of the images.

Color-based underwater object recognition using water light attenuation

In this article we present a new approach for object recognition in a robotic underwater context. Color is an attractive feature because of its simplicity and its robustness to scale changes, object positions and partial occlusions. Unfortunately, in the underwater medium, the colors are modified by attenuation and are not constant with the distance. To perform a color-based recognition of an object, we develop an algorithm robust with respect to the attenuation which takes into account the light modification during its path between the light source and the camera. Therefore, a given underwater object can be identified in an image by detecting all the colors compatible with its prior known color. Our method is fast, robust and needs a very few computers resources. We successfully used it when experimenting in the sea using a system we built. It is suitable for robotic applications where computers resources are limited and shared between various embedded devices. This novel concept enables the use of the color in many applications such as target interception, object tracking or obstacle detection.

AUV (Redermor) Obstacle Detection and Avoidance Experimental Evaluation

Autonomous underwater vehicles (AUV) are expected to perform survey missions in both known and unknown environments. While the primary mission of an AUV is data collection, generally achieved with a sidescan sonar or a multibeam echosounder, another key task is to guaranty its own security. This paper deals with the problem of obstacle detection and avoidance by means of a forward looking sonar (FLS) mounted on the GESMA Redermor experimental AUV.

Robust Multitarget Tracking in Forward-Looking Sonar Image Sequences Using Navigational Data

This paper presents a new approach to the problem of tracking objects in sequences of forward-looking sonar images. Unlike previous work, navigational data are taken as inputs to the state model of the Kalman filter used for tracking fixed obstacles. This model allows a robust prediction of their apparent motion in relation to the position of the sonar. A complete framework is presented where detection and data association issues are also discussed. An assessment of the proposed method has been carried out on real data from two different systems. Moreover, whereas the state model was first derived for a ground obstacle, a modified state model is proposed to estimate the altitude of the obstacle in relation to the sonar position using a number of successive pings.

Automatic Sea-Surface Obstacle Detection and Tracking in Forward-Looking Sonar Image Sequences

Automatic sea-surface object detection and tracking for safe autonomous underwater vehicle and submarine surfacing is a critical issue in relation to the accidents reported in the last decades. Here, we propose an efficient tool to detect and track sea-surface obstacles by processing forward-looking sonar images. The proposed method can detect either still or moving objects with and without wake. For each image sequence, a sequential procedure is proposed to detect various obstacle signatures. Then, target positions and velocities are estimated in Cartesian coordinates using the debiased converted measurement Kalman filter and the joint probabilistic data association filter. Detection and tracking stages exchange information in order to reduce the number of false alarms. Promising results are obtained using real data collected at sea with various objects and scenarios.

Model based classification of mine-like objects in sidescan sonar using the highlight information

This paper presents a model-based approach to perform underwater target classification. Very high resolution imaging sonar has increased the opportunities to use highlight information contained in the target acoustic signature whereas underwater target classification is still mainly based on the analysis of geometrical properties of the acoustic shadows. Supervised classifiers generally use experimental or simulated samples of target acoustic signature in the training stage but when the testing set is different from the training set the performance can be altered. Here the classification method consists in comparing the A-scan of the detected target with a set of simulated A-scans generated by our Sonar Image Simulator (SIS) in the same operational conditions. The used simulator relies on acoustical ray tracing techniques and takes into account complicated underwater physical process to simulate an accurate time response of underwater targets (A-scan). Practically the classifier is made of a cascade of matched filters. Each is built by simulating the A-scan for a given object in a given orientation (and /or for a given size). The resulting scores can be used to rank likelihood of belonging to object classes. The result is flexible and gives a percentage match for each class. With this approach the training set can be extended increasingly to improve classification when classes are strongly correlated. This classification process is assessed on a few real sidescan sonar data. These first results are finally discussed and further work is deduced to improve the general classification task.

Multisegmentation of sonar images using belief function theory

Today side scan sonar is one of the most efficient sensors for Rapid Environment Assessment missions. Unfortunately, features extracted from a given area are strongly dependent on the relative position of the sensor (e.g., due to the shadow or the gain variation). That could conduct to a bad segmentation of the seabed. However, due to the fact that operational systems give very often multiple views of the same area we use the redundancy. In this work, we propose to fuse multiview segmentations in order to outperform the seabed classification. First we present a way to characterize the seabed using as a start point, a texture analysis in order to extract parameters on images. Then, a classification method allows allocating a class according to the type of sediment for the different standpoints. The proposed classifier fusion is based on the belief function theory. We present results from a set of experiments conducted to evaluate the proposed approach with real sonar images and we discuss them.

Corrections to “Automatic Sea-Surface Obstacle Detection and Tracking in Forward-Looking Sonar Image Sequences”

Automatic sea-surface object detection and tracking for safe autonomous underwater vehicle and submarine surfacing is a critical issue in relation to the accidents reported in the last decades. Here, we propose an efficient tool to detect and track sea-surface obstacles by processing forward-looking sonar images. The proposed method can detect either still or moving objects with and without wake. For each image sequence, a sequential procedure is proposed to detect various obstacle signatures. Then, target positions and velocities are estimated in Cartesian coordinates using the debiased converted measurement Kalman filter and the joint probabilistic data association filter. Detection and tracking stages exchange information in order to reduce the number of false alarms. Promising results are obtained using real data collected at sea with various objects and scenarios.

Computing navigation corrections for co-registration of repeat-pass synthetic aperture sonar images

Interferometric processing of repeat-pass Synthetic Aperture Sonar (SAS) data requires a precise co-registration of the images in order to exploit the phase information for applications such as differential interferometry or coherent change detection. Co-registration of SAS images is made difficult by changing environmental conditions and positioning ambiguities that give rise to warping functions which are not well modeled by affine or polynomial transformations, resulting in non-negligible residual misregistration errors. The objective of this presentation will be to examine these warping functions in greater detail to gain a better physical understanding of the estimated along- and across-track displacements. A non-linear least-squares optimization is used to determine the corrections to be applied to the repeat-pass data that will yield a repeat-pass image that is co-registered with the original. These corrections include sway, heave and yaw as well as sound speed errors and element-level surge adjust...