Network Patterns of Zhongyuan Urban Agglomeration in China Based on Baidu Migration Data

As a new product of the Internet and big data era, migration data are of great significance for the revealing of the complex dynamic network patterns of urban agglomerations and for studying the relations between cities by using the “space of flows” model. Based on Baidu migration data of one week in 2021, this paper constructs a 30 × 30 rational data matrix for cities in Zhongyuan Urban Agglomeration and depicts the network pattern from static and dynamic perspectives by using social network analysis and dynamic network visualization. The results show that the network of Zhongyuan Urban Agglomeration is characterized by a circular structure with Zhengzhou as the center, a city belt around Zhengzhou as the connection, subcentral cities as the support and peripheral cities as the extension. Zhengzhou is the core city of the entire network, related to which the central and backbone networks divided in this paper account for nearly 40% of the total migration. Shangqiu, Luoyang, Zhoukou and Handan also play an important role in the structure of the migration network as subcentral cities. For a single city, the migration scale generally peaks on weekends and reaches its minimum during Tuesday to Thursday. Regarding the relations between cities, the migration variation can be divided into four types: peaking on Monday, peaking on weekends, bimodal and stable, and there are obvious phenomena of weekly commuting. In general, the links between cities outside Henan Province and other cities in the urban agglomeration are relatively weak, and the constraints of administrative regionalization on intercity migration are presumed to still exist. According to the results, the location advantage for multi-layer development and construction of Zhongyuan Urban Agglomeration should be made use of. In addition, the status as the core city and the radiation range should be strengthened, and the connections between the peripheral cities and the other cities should be improved, so as to promote the integrated and efficient development of the whole urban agglomeration.

On-Road Driver Emotion Recognition Using Facial Expression

With the development of intelligent automotive human-machine systems, driver emotion detection and recognition has become an emerging research topic. Facial expression-based emotion recognition approaches have achieved outstanding results on laboratory-controlled data. However, these studies cannot represent the environment of real driving situations. In order to address this, this paper proposes a facial expression-based on-road driver emotion recognition network called FERDERnet. This method divides the on-road driver facial expression recognition task into three modules: a face detection module that detects the driver’s face, an augmentation-based resampling module that performs data augmentation and resampling, and an emotion recognition module that adopts a deep convolutional neural network pre-trained on FER and CK+ datasets and then fine-tuned as a backbone for driver emotion recognition. This method adopts five different backbone networks as well as an ensemble method. Furthermore, to evaluate the proposed method, this paper collected an on-road driver facial expression dataset, which contains various road scenarios and the corresponding driver’s facial expression during the driving task. Experiments were performed on the on-road driver facial expression dataset that this paper collected. Based on efficiency and accuracy, the proposed FERDERnet with Xception backbone was effective in identifying on-road driver facial expressions and obtained superior performance compared to the baseline networks and some state-of-the-art networks.

Joint Classification and Regression for Visual Tracking with Fully Convolutional Siamese Networks

Visual tracking of generic objects is one of the fundamental but challenging problems in computer vision. Here, we propose a novel fully convolutional Siamese network to solve visual tracking by directly predicting the target bounding box in an end-to-end manner. We first reformulate the visual tracking task as two subproblems: a classification problem for pixel category prediction and a regression task for object status estimation at this pixel. With this decomposition, we design a simple yet effective Siamese architecture based classification and regression framework, termed SiamCAR, which consists of two subnetworks: a Siamese subnetwork for feature extraction and a classification-regression subnetwork for direct bounding box prediction. Since the proposed framework is both proposal- and anchor-free, SiamCAR can avoid the tedious hyper-parameter tuning of anchors, considerably simplifying the training. To demonstrate that a much simpler tracking framework can achieve superior tracking results, we conduct extensive experiments and comparisons with state-of-the-art trackers on a few challenging benchmarks. Without bells and whistles, SiamCAR achieves leading performance with a real-time speed. Furthermore, the ablation study validates that the proposed framework is effective with various backbone networks, and can benefit from deeper networks. Code is available at https://github.com/ohhhyeahhh/SiamCAR.

Skeleton-Based Attention Mask for Pedestrian Attribute Recognition Network

This paper presents an extended model for a pedestrian attribute recognition network utilizing skeleton data as a soft attention model to extract a local feature corresponding to a specific attribute. This technique helped keep valuable information surrounding the target area and handle the variation of human posture. The attention masks were designed to focus on the partial and the whole-body regions. This research utilized an augmented layer for data augmentation inside the network to reduce over-fitting errors. Our network was evaluated in two datasets (RAP and PETA) with various backbone networks (ResNet-50, Inception V3, and Inception-ResNet V2). The experimental result shows that our network improves overall classification performance with a mean accuracy of about 2–3% in the same backbone network, especially local attributes and various human postures.

Green distributed algorithm for energy saving in IP wired networks using sleep scheduling

Energy saving has become a critical issue and a great challenge in the past few decades, and a great effort as well is being made to reduce consumed energy. The Internet forms a major source for energy consumption. Therefore, in this work we propose an algorithm for energy saving in distributed backbone networks, the reduced energy consumption (RedCon) algorithm. In this paper, we introduce a new version for saving energy on the Internet by switching off underutilized links and switching on idle links when the network is overloaded in a distributed manner over the network nodes based on LSA messages and without any knowledge of the traffic matrix. Our algorithm is more accurate and outperforms other algorithms with its time checks and advanced learning algorithm.

Distributed Denial of Service Attacks Detection and Mitigation in Software Defined Mesh Networks

Networks configured in Mesh topology provide Network security in the form of redundancy of communication links. But redundancy also contributes to complexity in configuration and subsequent troubleshooting. Critical networks like Backbone Networks (used in Cloud Computing) deploy the Mesh topology which provides additional security in terms of redundancy to ensure availability of services. Distributed Denial of Service attacks are one of the most prominent attacks that cause an immense amount of loss of data as well as monetary losses to service providers. This paper proposes a method by which using SDN capabilities and sFlow-RT application, Distributed Denial of Service (DDoS) attacks is detected and consequently mitigated by using REST API to implement Policy Based Flow Management through the SDN Controller which will help in ensuring uninterrupted services in scenarios of such attacks and also further simply and enhance the management of Mesh architecture-based networks.

Deep Ensembles Based on Stochastic Activations for Semantic Segmentation

Semantic segmentation is a very popular topic in modern computer vision, and it has applications in many fields. Researchers have proposed a variety of architectures for semantic image segmentation. The most common ones exploit an encoder–decoder structure that aims to capture the semantics of the image and its low-level features. The encoder uses convolutional layers, in general with a stride larger than one, to extract the features, while the decoder recreates the image by upsampling and using skip connections with the first layers. The objective of this study is to propose a method for creating an ensemble of CNNs by enhancing diversity among networks with different activation functions. In this work, we use DeepLabV3+ as an architecture to test the effectiveness of creating an ensemble of networks by randomly changing the activation functions inside the network multiple times. We also use different backbone networks in our DeepLabV3+ to validate our findings. A comprehensive evaluation of the proposed approach is conducted across two different image segmentation problems: the first is from the medical field, i.e., polyp segmentation for early detection of colorectal cancer, and the second is skin detection for several different applications, including face detection, hand gesture recognition, and many others. As to the first problem, we manage to reach a Dice coefficient of 0.888, and a mean intersection over union (mIoU) of 0.825, in the competitive Kvasir-SEG dataset. The high performance of the proposed ensemble is confirmed in skin detection, where the proposed approach is ranked first concerning other state-of-the-art approaches (including HarDNet) in a large set of testing datasets.

Object Detectors’ Convolutional Neural Networks backbones : a review and a comparative study

Computer vision is a scientific field that deals with how computers can acquire significant level comprehension from computerized images or videos. One of the keystones of computer vision is object detection that aims to identify relevant features from video or image to detect objects. Backbone is the first stage in object detection algorithms that play a crucial role in object detection. Object detectors are usually provided with backbone networks designed for image classification. Object detection performance is highly based on features extracted by backbones, for instance, by simply replacing a backbone with its extended version, a large accuracy metric grows up. Additionally, the backbone's importance is demonstrated by its efficiency in real-time object detection. In this paper, we aim to accumulate the crucial role of the deep learning era and convolutional neural networks in particular in object detection tasks. We have analyzed and have been concentrating on a wide range of reviews on convolutional neural networks used as the backbone of object detection models. Building, therefore, a review of backbones that help researchers and scientists to use it as a guideline for their works.

CasTabDetectoRS: Cascade Network for Table Detection in Document Images with Recursive Feature Pyramid and Switchable Atrous Convolution

Table detection is a preliminary step in extracting reliable information from tables in scanned document images. We present CasTabDetectoRS, a novel end-to-end trainable table detection framework that operates on Cascade Mask R-CNN, including Recursive Feature Pyramid network and Switchable Atrous Convolution in the existing backbone architecture. By utilizing a comparativelyightweight backbone of ResNet-50, this paper demonstrates that superior results are attainable without relying on pre- and post-processing methods, heavier backbone networks (ResNet-101, ResNeXt-152), and memory-intensive deformable convolutions. We evaluate the proposed approach on five different publicly available table detection datasets. Our CasTabDetectoRS outperforms the previous state-of-the-art results on four datasets (ICDAR-19, TableBank, UNLV, and Marmot) and accomplishes comparable results on ICDAR-17 POD. Upon comparing with previous state-of-the-art results, we obtain a significant relative error reduction of 56.36%, 20%, 4.5%, and 3.5% on the datasets of ICDAR-19, TableBank, UNLV, and Marmot, respectively. Furthermore, this paper sets a new benchmark by performing exhaustive cross-datasets evaluations to exhibit the generalization capabilities of the proposed method.

Research on Object Detection Model Based on Feature Network Optimization

As the object detection dataset scale is smaller than the image recognition dataset ImageNet scale, transfer learning has become a basic training method for deep learning object detection models, which pre-trains the backbone network of the object detection model on an ImageNet dataset to extract features for detection tasks. However, the classification task of detection focuses on the salient region features of an object, while the location task of detection focuses on the edge features, so there is a certain deviation between the features extracted by a pretrained backbone network and those needed by a localization task. To solve this problem, a decoupled self-attention (DSA) module is proposed for one-stage object-detection models in this paper. A DSA includes two decoupled self-attention branches, so it can extract appropriate features for different tasks. It is located between the Feature Pyramid Networks (FPN) and head networks of subtasks, and used to independently extract global features for different tasks based on FPN-fused features. Although the DSA network module is simple, it can effectively improve the performance of object detection, and can easily be embedded in many detection models. Our experiments are based on the representative one-stage detection model RetinaNet. In the Common Objects in Context (COCO) dataset, when ResNet50 and ResNet101 are used as backbone networks, the detection performances can be increased by 0.4 and 0.5% AP, respectively. When the DSA module and object confidence task are both applied in RetinaNet, the detection performances based on ResNet50 and ResNet101 can be increased by 1.0 and 1.4% AP, respectively. The experiment results show the effectiveness of the DSA module.