In this paper we present a photogrammetry-based approach for deep-sea underwater surveys conducted from submarine and guided by knowledge-representation combined with logical (ontology). Two major issues are discussed in paper. The first concerns using photogrammetry submarine. Here the goal was to obtain set of images that completely covered selected site. Subsequently based on these images, low-resolution 3D model is obtained real-time, followed very high-resolution produced back...

In this article, we present an approach for a deep-sea survey based on photogrammetry using remotely operated underwater vehicle (ROV). A hybrid technique gives us real-time results, sufficient piloting the ROV from surface vessel and ensuring uniform coverage of site, as well recording high-definition images onboard computer that will later provide with millimetric precision. The measurements are made without any contact noninvasive. time required on-site is minimal corresponds to needed by...

Overexploitation is a major threat for the integrity of marine ecosystems. Understanding ecological consequences different extractive practices and mechanisms underlying recovery populations essential to ensure sustainable management plans. Precious corals are long-lived structural invertebrates, historically overfished, their conservation currently worldwide concern. However, processes poorly known. Here, we examined harvesting effects red coral Corallium rubrum by analyzing long-term...

The early detection of melanoma is the most efficient way to reduce its mortality rate. Dermatologists achieve this task with help dermoscopy, a non-invasive tool allowing visualization patterns skin lesions. Computer-aided diagnosis (CAD) systems developed on dermoscopic images are needed assist dermatologists. These rely mainly multiclass classification approaches. However, lesions by an automated system remains challenging task. Decomposing problem into binary can complexity initial and...

Gaze interpretation is a fundamental aspect of social communication, especially for children with autism spectrum disorder (ASD), who frequently encounter many difficulties in interactions. Despite the considerable advances made gaze tracking technologies, such as those based on RGB and RGB-D, accurate measurement direction remains significant scientific challenge. This paper proposes novel approach utilizing Azure Kinect to improve intention ASD, providing estimations both head position. To...

In this paper, we present a new method for counting people. This is based on the head detection after segmentation of human body by skeleton graph process. The silhouette computed and decomposed into set segments corresponding to head, torso limbs. structure captures minimal information about shape. No assumption made viewpoint, done pose Several results efficiency labelling process , particularly its structural properties heads within crowd. A proposed has been tested with an experiment...

Early detection of melanoma remains a daily challenge due to the increasing number cases and lack dermatologists. Thus, AI-assisted diagnosis is considered as possible solution for this issue. Despite great advances brought by deep learning especially convolutional neural networks (CNNs), computer-aided (CAD) systems are still not used in clinical practice. This may be explained dermatologist's fear being misled false negative assimilation CNNs "black box", making their decision process...

Melanoma is still a major health problem worldwide. Early diagnosis the first step toward reducing its mortality, but it remains challenge even for experienced dermatologists. Although computer-aided systems have been developed to help diagnosis, lack of insight into their predictions significant limitation acceptance by medical community. To tackle this issue, we designed handcrafted expert features representing color asymmetry within lesions, which are parts approach used dermatologists in...

ROV 3D project aims at developing innovative tools which link underwater photogrammetry and acoustic measurements from an active sensor. The results will be high resolution surveys of sites. new means methods developed aim reducing the investigation time in situ, proposing comprehensive non-intrusive measurement for studied environment. In this paper, we apply a pre-processing pipe line to increase SIFT SURF descriptors extraction quality order solve problem surveying archaeological wreck...

Since 1973 archaeology and computer science have developed close ties in Marseille. Two departments (computer archaeology) from the French National Centre for Scientific Research (CNRS) Marseille started working together laid cornerstone of Computer Applications Quantitative Methods Archaeology (CAA) community. also has advantage being located a very interesting place on Mediterranean Sea home to several famous laboratories, such as Cultural Heritage Department (DRASSM) or private companies...

Video-based person re-identification (Re-ID) is a challenging task aiming to match individuals across various cameras based on video sequences.While most existing Re-ID techniques focus solely appearance information, including gait could potentially improve systems.In this study, we propose, GAF-Net, novel approach that integrates with features for re-identifying individuals; the are extracted from RGB tracklets while skeletal pose estimation.These then combined into single feature allowing...

Light Sheet Fluorescence Microscopy (LSFM) has emerged as a valuable tool for neurobiologists, enabling the rapid and high-quality volumetric imaging of mice brains. However, inherent artifacts distortions introduced during process necessitate careful enhancement LSFM images optimal 3D reconstructions. This work aims to correct slice by before reconstructing volumes. Our approach involves three-step process: firstly, implementation deblurring algorithm using K. Becker; secondly, an automatic...

In this paper, we present a new method for counting people. This is based on the head detection after segmentation of human body by skeleton graph process. The silhouette computed and decomposed into set segments corresponding to , torso limbs. structure captures minimal information about shape. No assumption made viewpoint, done pose Several results efficiency labelling process particularly its structural properties heads within crowd. A proposed are evaluated crowd task in PETS 2010 dataset.

Since 1973 archeology and computer science have developed close ties in Marseille. Two departments (computer archaeology) from the French National Centre for Scientific Research (CNRS) Marseille started working together laid cornerstone of Computer Applications Quantitative Methods Archaeology (CAA) community. also has advantage being located a very interesting place on Mediterranean Sea home to several famous laboratories, such as Cultural Heritage Department (DRASSM) or private companies...

In this paper, we present a new person re-identification method based on appearance classification and silhouette part segmentation. crowded areas, heads are considered as most apparent parts, hence the typical advantage of using skeleton graph for head detection location people after partial occlusion. The consists in characterizing into two classes, frontal back appearance, detector orthogonal iteration algorithm pose estimation. segmentation divides three horizontal ideally corresponding...

The people's counting is one of the most important parts in design any system for behavioral analysis.It used to measure and manage flow within zones with restricted attendance.In this work, we propose a strategy number people entering leaving given closed area.Our method based on use depth map obtained from Kinect sensor installed zenithal position respect motion direction.It count crossing virtual line interest (LOI).The proposed two main modules detection module that detect individuals...

Due to its increasing incidence, skin cancer, and especially melanoma, is considered a major public health issue. Manually detecting lesions (SL) from dermoscopy images difficult time-consuming process. Thus, researchers designed computer-aided diagnosis (CAD) systems assist dermatologists in the early detection of cancer. Moreover, SL naturally exhibits long-tailed distribution due complex patient-level conditions existence rare diseases. Very limited research for handling this issue exists...

This article describes the set of photogrammetric tools developed for monitoring Mediterranean red coral Corallium rubrum populations. The description encompasses full processing chain: from image acquisition to information extraction and data interpretation. methods applied take advantage existing new, innovative specific developments in order acquire on relevant ecological concerning structure functioning a population. presented here are based on: (i) automatic orientation using coded...

This paper provides details of hardware and software conception realization a stereo embedded system for underwater imaging. The several functions that facilitate surveys run smoothly in real-time. A first post-image acquisition module direct visual feedback on the quality taken images which helps appropriate actions to be regarding movement speed lighting conditions. Our main contribution is light odometry method adapted context. proposed uses captured image stream provide real-time...

Autism Spectrum Disorders (ASD) is a developmental disorder that affects communications, social skills or behaviours can occur in some people. Children adults with ASD often have repetitive motor movements unusual behaviours. The objective of this work to automatically detect stereotypical real time using Kinect sensor. approach based on the $P Point-Cloud Recogniser identify multi-stroke gestures as point clouds. This paper presents new methodology five movements: body rocking, hand...