Djamal Merad

ROV-3D: 3d underwater survey combining optical and acoustic sensor

ROV 3D project aims at developing innovative tools which link underwater photogrammetry and acoustic measurements from an active underwater sensor. The results will be 3D high resolution surveys of underwater sites. The new means and methods developed aim at reducing the investigation time in situ, and proposing comprehensive and non-intrusive measurement tools for the studied environment. In this paper, we are presenting a new method of 3D surveys which are dedicated to high resolution modeling of underwater sites. The main met constraints in situ are taken into account and this method leads to a precise 3D reconstruction. Some examples will present both the main obtained results and their limitations. We will end with the perspectives and the necessary improvements to the method, so as to automate the multimodal registration step.

An information system for medieval archaeology based on photogrammetry and archaeological database: the Shawbak castle project

The paper presents an interdisciplinary project which is a work in progress towards a 3D Geographical Information System (GIS) dedicated to Cultural Heritage with a specific focus application on the Castle of Shawbak, one of the best preserved rural medieval settlements in the entire Middle East). The Shawbak archaeological project is a specific and integrated project between medieval archaeological research and computer vision done thanks to a long cooperation between University of Florence and CNRS, LSIS, Marseille. Focusing mainly on stratigraphical analysis of upstanding structures we provide archaeologists with two-step pipeline. First a survey process using photogrammetry, both in a traditional way with additional annotations and using the most advanced technique to obtain dense maps and then a tool for statistical analysis. Two main applications are presented here, stratigraphy analysis with Harris matrix computed on the fly from the 3D viewer and statistical tools, clustering operation on ashlar in order to show new relationships between the measured artifacts. All these developments are written in Java within Arpenteur framework[4].

Head detection based on skeleton graph method for counting people in crowded environments

Abstract. We describe a method for detecting heads in order to count people in crowded environments using a single camera. The main difference between this method and traditional ones consists of adapting skeleton graph analysis techniques for distinguishing individuals in crowded environments. First, a graph skeleton is calculated for each selected blob in a scene after having performed motion estimation. Then, the structural property of each blob is explored to detect possible heads in order to estimate the number of people. Each detected head in the skeleton silhouette is identified as being in an independent or partial occlusion state and is updated during a tracking process. Finally, the results of our experiments are presented to demonstrate the robustness of our method.

The ROV 3D Project: Deep-Sea Underwater Survey Using Photogrammetry: Applications for Underwater Archaeology

In this article, we present an approach for a deep-sea survey based on photogrammetry using a remotely operated underwater vehicle (ROV). A hybrid technique gives us real-time results, sufficient for piloting the ROV from the surface vessel and ensuring a uniform coverage of the site, as well as recording high-definition images using an onboard computer that will later provide a survey with millimetric precision. The measurements are made without any contact and are noninvasive. The time required on-site is minimal and corresponds to the time needed by the ROV to cover the zone. With the photos taken at a frame rate synchronized at 10Hz, the ROV required 2 hours to perform the experiment presented in this article: the survey of the Roman shipwreck Cap Bénat 4, at a depth of 328m. The approach presented in this work was developed in the scope of the ROV 3D project. This project, financed by the Fond Unique Interministériel (FUI) for 3 years, brings together two industrial partners and a research laboratory. Companie Maritime d’Expertise (COMEX) coordinated this project.

Partial 3d-object retrieval using level curves

3D object recognition from 3D scenes, is one of the challenges of several researchers in the field of computer vision, engineering and Robotics. The occlusion is one of the problems that we can found. One of the possible solutions in this situation is to find a part of an object in the scene that can be identified. For this reason, we are mainly interested to partial shape retrieval methods. In this paper, we present a new approach for 3D partial object retrieval based on level curves matching. Our approach can be used as an alternative solution for classification-based methods. We generate in the off-line step a dataset by using a viewing sphere to extract levels curves at different points of view. The level curves are a set of 2D planar contours that are the projection of points on several perpendicular planes. The level curves of each query partial object is compared with a set of level curves that define one 3D object from the dataset. The number of matched curves between a partial object and complete object represent the weight of that class. The class with the heavy weight is identify as the class of the query object.

Underwater Multimodal Survey: Merging Optical and Acoustic Data

ROV 3D project aims at developing innovative tools which link underwater photogrammetry and acoustic measurements from an active underwater sensor. The results will be 3D high resolution surveys of underwater sites and landscapes useful to keep in memory cultural and natural heritage. The new means and methods developed aim at reducing the investigation time in situ, and proposing comprehensive and non-intrusive measurement tools for the studied environment.

Photogrammetry for medieval archaeology: A way to represent and analyse stratigraphy

The paper presents an interdisciplinary project which is a work in progress towards a 3D Geographical Information System (GIS) dedicated to Cultural Heritage with a specific focus application on the Castle of Shawbak, also known as the “Crac de Montréal”, one of the best preserved rural medieval settlements in the entire Middle East. The Shawbak archaeological project is a specific and integrated project between medieval archaeological research and computer vision done thanks to a long cooperation between University of Florence and CNRS. Focusing mainly on stratigraphical analysis of upstanding structures we provide archaeologists a two-step pipeline. First a survey process using photogrammetry, both in a traditional way and using the most advanced technique for obtained dense map and then a tool for statistical analysis. The photogrammetric survey is driven by an archaeological knowledge which is formalized by ontologies as a link between all the archaeological concepts which are surveyed. The archaeological knowledge studied is now limited to stratigraphy of upstanding structure using a stone by stone survey as well as a 3D reprojection of archaeologist design made on photographs. The 3D GIS is the last step of this chain and aims the automatic production of 3D models through archaeological database queries: these 3D models are in fact a graphical image of the database and at the same time the interface through which the user is able to modify it and produce different kind of analyzing. All these developments are written in Java within ARPENTEUR framework.

People Counting based on Kinect Depth Data.

The people’s counting is one of the most important parts in the design of any system for behavioral analysis. It is used to measure and manage people’s flow within zones with restricted attendance. In this work, we propose a counting strategy for counting the number of people entering and leaving a given closed area. Our counting method is based on the use of depth map obtained from a Kinect sensor installed in a zenithal position with respect to the motion direction. It is used to count the number of people crossing a virtual line of interest (LOI). The proposed method is based on the use of two main modules a people detection module that is used to detect individuals crossing the LOI and a tracking module that is used to track detected individuals to determine the direction of their motions. The people detection is based on the design of a smart sensor that is used with both the grayscale image that represents depth changes and the binary image that represents foreground objects within the depth map to detect and localize individuals. Then, these individuals are tracked by the second module to determine the direction of their motions.

Fast Visual Odometry for a Low-Cost Underwater Embedded Stereo System †

This paper provides details of hardware and software conception and realization of a stereo embedded system for underwater imaging. The system provides several functions that facilitate underwater surveys and run smoothly in real-time. A first post-image acquisition module provides direct visual feedback on the quality of the taken images which helps appropriate actions to be taken regarding movement speed and lighting conditions. Our main contribution is a light visual odometry method adapted to the underwater context. The proposed method uses the captured stereo image stream to provide real-time navigation and a site coverage map which is necessary to conduct a complete underwater survey. The visual odometry uses a stochastic pose representation and semi-global optimization approach to handle large sites and provides long-term autonomy, whereas a novel stereo matching approach adapted to underwater imaging and system attached lighting allows fast processing and suitability to low computational resource systems. The system is tested in a real context and shows its robustness and promising future potential.

People tracking in multi-camera systems: a review

Ubiquitousness of multiple cameras in surveillance systems is very beneficial for studying peoples behavior. The multiple views of the observed scene permit the management of dynamic occlusions and failures that may affect any sensor. The multi-camera tracking of objects is considered as the basic step in the design of intelligent surveillance applications. This thematic had been addressed in several researches. Various methods had been proposed to achieve an accurate tracking in the most challenging conditions as occlusions and lighting variations. These methods are addressed in two main research lines: the centralized and the distributed tracking approaches. In this paper, we propose an overview of the multi-camera tracking of objects which summarizes and classifies the most used existing methods.