Internet of Things (IoT) has recently received a great attention due to its potential and capacity to be integrated into any complex system. As a result of rapid development of sensing technologies such as radio-frequency identification, sensors and the convergence of information technologies such as wireless communication and Internet, IoT is emerging as an important technology for monitoring systems. This paper reviews and introduces a framework for structural health monitoring (SHM) using IoT technologies on intelligent and reliable monitoring. Specifically, technologies involved in IoT and SHM system implementation as well as data routing strategy in IoT environment are presented. As the amount of data generated by sensing devices are voluminous and faster than ever, big data solutions are introduced to deal with the complex and large amount of data collected from sensors installed on structures.

Structural Health Monitoring Framework Based on Internet of Things: A Survey

Transformer, first applied to the field of natural language processing, is a type of deep neural network mainly based on the self-attention mechanism. Thanks to its strong representation capabilities, researchers are looking at ways to apply transformer to computer vision tasks. In a variety of visual benchmarks, transformer-based models perform similar to or better than other types of networks such as convolutional and recurrent neural networks. Given its high performance and less need for vision-specific inductive bias, transformer is receiving more and more attention from the computer vision community. In this paper, we review these vision transformer models by categorizing them in different tasks and analyzing their advantages and disadvantages. The main categories we explore include the backbone network, high/mid-level vision, low-level vision, and video processing. We also include efficient transformer methods for pushing transformer into real device-based applications. Furthermore, we also take a brief look at the self-attention mechanism in computer vision, as it is the base component in transformer. Toward the end of this paper, we discuss the challenges and provide several further research directions for vision transformers. 

/pdf/a-survey-on-vision-transformer-3fin6y51.pdf

A Survey on Vision Transformer

Computational saliency models for still images have gained significant popularity in recent years. Saliency prediction from videos, on the other hand, has received relatively little interest from the community. Motivated by this, in this paper, we study the use of deep learning for dynamic saliency prediction and propose the so-called spatio-temporal saliency networks. The key to our models is the architecture of two-stream networks where we investigate different fusion mechanisms to integrate spatial and temporal information. We evaluate our models on the dynamic images and eye movements and University of Central Florida-Sports datasets and present highly competitive results against the existing state-of-the-art models. We also carry out some experiments on a number of still images from the MIT300 dataset by exploiting the optical flow maps predicted from these images. Our results show that considering inherent motion information in this way can be helpful for static saliency estimation.

Spatio-Temporal Saliency Networks for Dynamic Saliency Prediction

In the last few years, the Internet of Things (IoT) has proved to be an interesting and promising paradigm that aims to contribute to countless applications by connecting more physical “things” to the Internet. Although it emerged as a major enabler for many next-generation applications, it also introduced new challenges to already saturated networks. The IoT is already coming to life especially in healthcare and smart environment applications adding a large number of low-powered sensors and actuators to improve lifestyle and introduce new services to the community. The Internet Engineering Task Force (IETF) developed RPL as the routing protocol for low-power and lossy networks (LLNs) and standardized it in RFC6550 in 2012. RPL quickly gained interest, and many research papers were introduced to evaluate and improve its performance in different applications. In this paper, we present a discussion of the main aspects of RPL and the advantages and disadvantages of using it in different IoT applications. We also review the available research related to RPL in a systematic manner based on the enhancement area and the service type. In addition to that, we compare related RPL-based protocols in terms of energy efficiency, reliability, flexibility, robustness, and security. Finally, we present our conclusions and discuss the possible future directions of RPL and its applicability in the Internet of the future.

/pdf/rpl-based-routing-protocols-in-iot-applications-a-review-4uibgx51o1.pdf

RPL-Based Routing Protocols in IoT Applications: A Review

In this paper, we propose a novel method to discover co-salient objects from a group of images, which is modeled as a linear fusion of an intra-image saliency (IaIS) map and an inter-image saliency (IrIS) map. The first term is to measure the salient objects from each image using multiscale segmentation voting. The second term is designed to detect the co-salient objects from a group of images. To compute the IrIS map, we perform the pairwise similarity ranking based on an image pyramid representation. A minimum spanning tree is then constructed to determine the image matching order. For each region in an image, we design three types of visual descriptors, which are extracted from the local appearance, e.g., color, color co-occurrence and shape properties. The final region matching problem between the images is formulated as an assignment problem that can be optimized by linear programming. Experimental evaluation on a number of images demonstrates the good performance of the proposed method on co-salient object detection.

/pdf/co-salient-object-detection-from-multiple-images-33eenx9wti.pdf

Co-Salient Object Detection From Multiple Images

Designing routing protocols in Low power and Lossy Networks (LLNs) imposes great challenges. In emergency scenarios, the large and rapid data traffic caused by emergencies will lead to network congestion and bring about significant packet loss and delay. Routing protocol for LLNs (RPL) is the IETF standard for IPv6 routing in LLNs. The basic version of RPL uses Expected Transmission Count (ETX) as the default routing metric; it cannot solve the problem of sudden large data traffic. In this paper, we propose a congestion avoidance multipath routing protocol which uses composite routing metrics based on RPL, named CA-RPL. A routing metric for RPL that minimized the average delay towards the DAG root is proposed, and the weight of each path is computed by four metrics. The mechanism is explained and its performance is evaluated through simulation experiments based on Contiki. Simulation results show that the proposed CA-RPL reduces the average time delay by about 30% compared to original RPL when the interpacket interval is short and has almost 20% reduction in packet loss ratio. The CA-RPL can effectively alleviate the network congestion in the network with poor link quality and large data traffic and significantly improve the performance of LLNs.

/pdf/toward-improved-rpl-a-congestion-avoidance-multipath-routing-mstxgpxojk.pdf

Toward Improved RPL: A Congestion Avoidance Multipath Routing Protocol with Time Factor for Wireless Sensor Networks

In this paper, we propose a new algorithm for image re-ranking in web image search applications. The proposed method focuses on investigating the following two mechanisms: 1) Visual consistency. In most web image search cases, the images that closely related to the search query are visually similar. These visually consistent images which occur most frequently in the first few web pages will be given higher ranks. 2) Visual saliency. From visual aspect, it is obvious that salient images would be easier to catch users' eyes, and it is observed that these visually salient images in the front pages are often relevant to the user's query. By integrating the above two mechanisms, our method can efficiently re-rank the images from search engines and obtain a more satisfactory search result. Experimental results on a real-world web image dataset demonstrate that our approach can effectively improve the performance of image retrieval.

Integrating Visual Saliency and Consistency for Re-Ranking Image Search Results

Goal-driven top-down mechanism plays important role in the case of object detection and recognition. In this paper, we propose a top-down computational model for goal-driven saliency detection based on the coding-based classification framework. It consists of four successive steps: feature extraction, descriptor coding, contextual pooling and saliency prediction. Particularly, we investigate the effect of spatial context information for saliency detection, and propose a block-wise spatial pooling operation to take advantage of contextual cues in multiple neighborhood scales and orientations. The experimental results on three datasets demonstrate that our method can effectively exploit contextual information and achieves the state-of-the-art performance on top-down saliency detection task.

Top-Down Saliency Detection via Contextual Pooling

Based on pattern uniformity measure and the number of ones in the Local Binary Pattern (LBP) codes, this paper proposes an Improved Local Binary Pattern (ILBP) operator to describe local image texture more effectively. The ILBP operator discovers an important group of basic primitives such as lines, T-junctions, and cross-intersections, which are ignored by uniform LBP operator. Such local primitives are as crucial as those represented by uniform patterns for recognition tasks. The resulting ILBP feature is more discriminative than traditional LBP feature although they are both invariant in terms of monotonic gray-scale variation and rotation transformation.

/pdf/an-improved-local-binary-pattern-operator-for-texture-3fzk5cstde.pdf

An improved local binary pattern operator for texture classification

This paper aims to calibrate the mounted parameters between the LIDAR and the motor in a low-cost 3D LIDAR device. It proposes the model of the aimed 3D LIDAR device and analyzes the influence of all mounted parameters. The study aims to find a way more accurate and simple to calibrate those mounted parameters.,This method minimizes the coplanarity and area of the plane scanned to estimate the mounted parameters. Within the method, the authors build different cost function for rotation parameters and translation parameters; thus, the parameter estimation problem of 4-degree-of-freedom (DOF) is decoupled into 2-DOF estimation problem, achieving the calibration of these two types of parameters.,This paper proposes a calibration method for accurately estimating the mounted parameters between a 2D LIDAR and rotating platform, which realizes the estimation of 2-DOF rotation parameters and 2-DOF translation parameters without additional hardware.,Unlike previous plane-based calibration techniques, the main advantage of the proposed method is that the algorithm can estimate the most and more accurate parameters with no more hardware.

Jun Huang

Papers

Toward Improved RPL: A Congestion Avoidance Multipath Routing Protocol with Time Factor for Wireless Sensor Networks

Integrating Visual Saliency and Consistency for Re-Ranking Image Search Results

Top-Down Saliency Detection via Contextual Pooling

An improved local binary pattern operator for texture classification

Calibration of rotating 2D LIDAR based on simple plane measurement