The arrival of Bluetooth Low Energy (BLE) creates opportunities for great innovations. One possible application is indoor localization. This paper presents a system that can track mobile device and help finding their location within a building perimeter. With the help of BLE beacons that can be deployed in different locations, the position of a mobile device can be estimated using RSSI techniques and trilateration methods. The overall system is controlled using MQTT protocol. Along this paper, we describe our system, the techniques we use, and the experiments we conducted along with the results.

Watershed segmentation generally produces a severe over-segmentation. Marker based segmentation creates watershed partitions in which each region contains a marker, avoiding the oversegmentation. We obtain a coherent picture of marker based segmentation by modelling it on node or edge weighted graphs. Links are established between the methods published in the literature and some misconceptions are corrected.

By 2020, more than 50 billion devices will be connected through radio communications. In conjunction with the rapid growth of the Internet of Things (IoT) market, low power wide area networks (LPWAN) have become a popular low-rate long-range radio communication technology. Sigfox, LoRa, and NB-IoT are the three leading LPWAN technologies that compete for large-scale IoT deployment. This paper provides a comprehensive and comparative study of these technologies, which serve as efficient solutions to connect smart, autonomous, and heterogeneous devices. We show that Sigfox and LoRa are advantageous in terms of battery lifetime, capacity, and cost. Meanwhile, NB-IoT offers benefits in terms of latency and quality of service. In addition, we analyze the IoT success factors of these LPWAN technologies, and we consider application scenarios and explain which technology is the best fit for each of these scenarios.

Cloud computing was considered last decade as a principal support for Internet of Things (IoT) applications. However, cloud solution introduces various constraints including mobility, reliable, scalability, and power consumption. To face such issues, researchers have proposed several solutions including mobile cloud computing, fog computing, and smart gateways. Recently, a PSS (product-service system) concept was proposed in the literature. In this paper, we focused on this concept which exploits the IoT object resources to further avoid IoT architecture issues while still meeting low-power and small-size object requirements. Thus, this paper discusses the emerging technologies and techniques that would be exploited towards PSS achievement. In addition, the promising hardware platforms for PSS achievement are presented and compared in terms of energy consumption.

We propose a framework for image characterization using hierarchies of segmentations. For this purpose, we structure the space of hierarchies using the Gromov-Hausdorff distance. We propose different ways of combining hierarchies and study their properties thanks to the GH distance. We then expose how to leverage the combinatorial space of hierarchies to derive efficient image representations. This framework opens a path for a controlled exploration and use of the combinatorial space of hierarchies. (C) 2019 Elsevier B.V. All rights reserved.

Indoor positioning is nowadays an emerging and mandatory technology for Internet of Things. Various solutions have been proposed in the literature to achieve high localization accuracy. However, these existing solutions are cloud-based, thus, all the computing and intelligence is implemented in the Cloud. In this paper, we present a new positioning system which embedded all the computing in the mobile device. The proposed solution achieves lower energy consumption while still offering high accuracy of 1.5 meters within an industrial building area. To this end, BLE beacons, RSSI, and trilateration techniques are exploited. Moreover, mobile devices are controlled using MQTT protocol. All advantages of our contribution are presented using various real experiments in an industrial building.

LPWAN are actually the most popular low cost, long battery lifetime, and long range communication technology for IoT applications. This paper presents a comprehensive and comparative study on three actually leading LPWAN technologies called Sigfox, LoRaWAN, and NB-IoT. We show that Sigfox and LoRaWAN excel on network capacity, devices lifetime, and cost. Whereas, NB-IoT excels on quality of service and latency. In addition, we consider application scenarios and explain which technology fits best to guide future researchers and industrials.

This chapter presents the principles of morphological segmentation. Segmentation is one of the key problems in image processing. In fact, one should say segmentations because there exist as many techniques as there are specific situations. An original method of segmentation based on the use of watershed lines has been developed in the framework of mathematical morphology. The chapter describes some useful morphological tools for segmentation: gradient, top-hat transform, distance function, geodesic distance function, and geodesic reconstructions. The gradient image is used in the watershed transformation, because the main criterion for the segmentation in many applications is the homogeneity of the gray values of the objects present in the image. The problems encountered in the segmentation process will be best illustrated by presenting a complete and typical segmentation problem in the field of automated cytology. The oversegmentation produced by direct construction of the watershed line is due to the fact that every regional minimum becomes the center of a catchment basin.

A hierarchy is a series of nested partitions in which a coarser partition results from merging regions of finer ones. Each hierarchy derived from an image provides a particular structural description of the image content, depending upon the criteria for merging neighboring regions. Distinct hierarchies derived from a same image reflect its various facets and the distances between them nicely characterize its content. In this paper the hierarchies are constructed with the versatile stochastic watershed algorithm and their inter-distances are measured with the Gromov-Hausdorff distance. Experiments conducted on images simulated by dead leaves model illustrate the advantages of our approach in terms of learning efficiency and understandability of the results.

Image segmentation is the process of partitioning an image into a set of meaningful regions according to some criteria. Hierarchical segmentation has emerged as a major trend in this regard as it favors the emergence of important regions at different scales. On the other hand, many methods allow us to have prior information on the position of structures of interest in the images. In this paper, we present a versatile hierarchical segmentation method that takes into account any prior spatial information and outputs a hierarchical segmentation that emphasizes the contours or regions of interest while preserving the important structures in the image. An application of this method to the weakly-supervised segmentation problem is presented.

Focusing on catchment zones which may overlap rather than on catchment basins which do not overlap allows to devise innovative segmentation strategies.

Image segmentation is the process of partitioning an image into a set of meaningful regions according to some criteria. Hierarchical segmentation has emerged as a major trend in this regard as it favors the emergence of important regions at different scales. On the other hand, many methods allow us to have prior information on the position of structures of interest in the images. In this paper, we present a versatile hierarchical segmentation method that takes into account any prior spatial information and outputs a hierarchical segmentation that emphasizes the contours or regions of interest while preserving the important structures in the image. Several applications are presented that illustrate the method versatility and efficiency.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. A short tour of mathematical morphology on edge and vertex weighted graphs Laurent Najman, Fernand Meyer

The segmentation, seen as the association of a partition with an image, is a difficult task. It can be decomposed in two steps: at first, a family of contours associated with a series of nested partitions (or hierarchy) is created and organized, then pertinent contours are extracted. A coarser partition is obtained by merging adjacent regions of a finer partition. The strength of a contour is then measured by the level of the hierarchy for which its two adjacent regions merge. We present an automatic segmentation strategy using a wide range of stochastic watershed hierarchies. For a given set of homogeneous images, our approach selects automatically the best hierarchy and cut level to perform image simplification given an evaluation score. Experimental results illustrate the advantages of our approach on several real-life images datasets.

We study and present two new algorithms for constructing the waterfall hierarchy of a topographic surface. The first models a topographic surface as a flooding graph, each node representing a lake filling a catchment basin up to its lowest pass point ; each edge representing such a pass point. The second algorithm produces the waterfall partition in one pass through the edges of a minimum spanning tree of the region adjacency graph associated to a topographic surface.

In 1991 I described a particularly simple and elegant watershed algorithm, where the flooding a topographic surface was scheduled by a hierarchical queue. In 2004 the watershed line has been described as the skeleton by zone of influence for the topographic distance. The same algorithm still applies. In 2012 I defined a new distance based on a lexicographic ordering of the downstream paths leading each node to a regional minimum. Without changing a iota, the same algorithm does the job.

We present a segmentation strategy which first constructs a hierarchy, i.e. a series of nested partitions. A coarser partition is obtained by merging adjacent regions in a finer partition. The strength of a contour is then measured by the level of the hierarchy for which its two adjacent regions merge. Various strategies are presented for constructing hierarchies which highlight specific features of the image. The last part shows how the hierarchies lead to a final segmentation.

This paper presents new structural statistical matrices which are gray level size zone matrix (SZM) texture descriptor variants. The SZM is based on the cooccurrences of size/intensity of each flat zone (connected pixels with the same gray level). The first improvement increases the information processed by merging multiple gray-level quantizations and reduces the required parameter numbers. New improved descriptors were especially designed for supervised cell texture classification. They are illustrated thanks to two different databases built from quantitative cell biology. The second alternative characterizes the DNA organization during the mitosis, according to zone intensities radial distribution. The third variant is a matrix structure generalization for the fibrous texture analysis, by changing the intensity/size pair into the length/orientation pair of each region.