Search

In order to realize the automatic dispensing and remote monitoring of emulsions in the coal mine production process and improve the efficiency and safety of coal mine production, an automatic dispensing and remote monitoring system of emulsions based on the IoT platform was built. The APRUS X industrial network management adapter in this system is the upper computer of Siemens S7-200SMART PLC and the data transmission hub between the dispensing system and the MixIOT cloud platform, while the MixIOT cloud platform realizes the functions of data visualization and remote control terminal; the PLC acts as the control core and monitors the weight sensors, solenoid valves, DC brushless booster pumps, agitators and other equipment at the same time. The data collected by the sensors are uploaded to the cloud through IOT technology and combined with real-time data to realize automatic dispensing and remote control of emulsion. Meanwhile, remote monitoring is realized through Web page and apps, which is convenient for managers to monitor the production site in real time and realize intelligent production. There are 27 groups of liquid dispensing experiments, and the average relative error of emulsion concentration ratio obtained after 27 groups of tests is 4.8%, and the accuracy requirement meets the actual engineering needs. The research results are of great significance to promote the intelligent production of coal mines and the application of industrial IOT technology, and provide specific reference and references for the research and application in related fields.

The road segmentation task is to extract the road surface from the image at pixel level. In road segmentation for remote sensing images, deep learning-based methods have shown high-quality results in various scenarios. However, existing segmentation methods usually produce discontinuous roads, which is not beneficial to applying practical scenarios. We propose a multi-task learning method of road segmentation, direction estimation and road edge learning to make our model connect roads reasonably. Moreover, we use the initial road segmentation results and the direction estimation results to make cascade inference to improve the model’s performance. We adopt the Canny operator to extract the edge information of images as the auxiliary modality fusion. We demonstrate our method’s effectiveness on two large-scale road segmentation datasets DeepGlobe and SpaceNet.

With sufficient practice, humans can grab objects they have never seen before through brain decision-making. However, the manipulators, which has a wide range of applications in industrial production, can still only grab specific objects. Because most of the grasp algorithms rely on prior knowledge such as hand-eye calibration results, object model features, and can only target specific types of objects. When the task scenario and the operation target change, it cannot perform effective redeployment. In order to solve the above problems, academia often uses reinforcement learning to train grasping algorithms. However, the method of reinforcement learning in the field of manipulators grasping mainly encounters these main problems: insufficient sample utilization, poor algorithm stability, and limited exploration. This article uses LfD, BC, and DDPG to improve sample utilization. Use multiple critics to integrate and evaluate input actions to solve the problem of algorithm instability. Finally, inspired by Thompson's sampling idea, the input action is evaluated from different angles, which increases the algorithm's exploration of the environment and reduces the number of interactions with the environment. EDDPG and EBDDPG algorithm is designed in the article. In order to further improve the generalization ability of the algorithm, this article does not use extra information that is difficult to obtain directly on the physical platform, such as the real coordinates of the target object and the continuous motion space at the end of the manipulator in the Cartesian coordinate system is used as the output of the decision. The simulation results show that, under the same number of interactions, the manipulators' success rate in grabbing 1000 random objects has increased more than double and reached state-of-the-art(SOTA) performance.

In air cushion furnace, the floating height is a key process parameter which greatly affects the quality and production efficiency of high quality metal strips. However, the floating height is hard to be collected in the complex and abominable industry environment. Furthermore, due to the flow field characteristics, some important process variables are difficult to accurately calculate by traditional mechanism modeling methods. In order to accurately predict the floating height, firstly, a low discrepancy heuristic evolution ELM and ground effect theory based serial hybrid soft sensor model is proposed, which constituted by a mechanism model and two data driven models. Secondly, based on the force equilibrium equation and ground effect theory, the mechanism model is constructed, which describes the relationship between the floating height and the process variables including the jet impact angle. Thirdly, a low discrepancy heuristic evolution ELM is proposed as the data driven model to predict the jet impinging angle. In the data driven model, the novel dual mutation strategies collaboration differential evolution is proposed to guarantee the low discrepancy and physical applicability of data driven model. The effectiveness of the proposed method was validated on the self-developed air cushion experiment platform and got desirable experimental results. The research lays an important foundation for the successful implementation of monitoring and control of the strip floating process.

Network management is facing a great challenge to analyze and identify encrypted network traffic with specific applications and protocols. A significant number of network users applying different encryption techniques to network applications and services to hide the true nature of the network communication. These challenges attract the network community to improve network security and enhance network service quality. Network managers need novel techniques to cope with the failure and shortcomings of the port-based and payload-based classification methods of encrypted network traffic due to emergent security technologies. Mainly, the famous network hopping mechanisms used to make network traffic unknown and anonymous are VPN (virtual private network) and TOR (Onion Router). This paper presents a novel scheme to unveil encrypted network traffic and easily identify the tunneled and anonymous network traffic. The proposed identification scheme uses the highly desirable deep learning techniques to easily and efficiently identify the anonymous network traffic and extract the Voice over IP (VoIP) and Non VoIP ones within encrypted traffic flows. Finally, the captured traffic has been classified into four different categories, i-e., VPN VoIP, VPN Non-VoIP, TOR VoIP, and TOR Non-VoIP. The experimental results show that our identification engine is extremely robust to VPN and TOR network traffic.

Due to the capability of passing through heavily censored networks or gateway equipped with the traffic-monitoring module, DNS tunnel has been the dominant covert communication technique for command and control between the victim and the attacker in network attack events. Although the discovery of DNS tunnel has been intensively studied, the internal application behavior identification for DNS tunnels still remains a challenging problem. The fine-gained identification can help to reveal more behavior information wrapped in DNS tunnels. In this study, we investigate the spatial-temporal information from the raw packets to identify the internal application behaviors in DNS tunnels. Multi-dimensional features on packet length and timing for DNS tunnels with different internal application behaviors are incorporated with a machine-learning algorithm to identify the internal application behaviors in DNS tunnels. We consider 4 common types of application behaviors in our research, including browsing webpages, emailing, downloading data, and controlling the remote servers. The experimental results show that the proposed scheme can achieve higher identification accuracy with a much lower packet consuming rate when compared with the state-of-the-art internal protocol identification scheme. The experiment results depict that our proposed scheme is better in terms of F-score, which can reach 99% with only 100 packets.

As one of the most critical infrastructure, the power grid has been increasingly threatened by network attacks, especially advanced persistent threats (APTs). APT in the power grid is a continual and stealthy attack that analyzes the interaction between the cyber layer and the physical layer. The existing offensive and defensive processes for power grid using honeypots against APTs are modeled based on full rationality. Therefore, both the attacker and the defender make decisions to maximize their payoffs under full rationality. However, fully rational decisions made by end-users are not always conformed with the real cases, and prospect theory is a typical boundedly rational method to model these deviations. In this study, we propose a subjective APT-honeypot game model to study the offensive and defensive interactions between the attacker and the defender based on the prospect theory. In this model, we protect the power grid bus nodes by deploying honeypots, which consider both low- and high-interaction honeypot modes. We prove the existence of Bayesian-Nash equilibrium strategies in defense and attack strategies under bounded rationality. In addition, we used IEEE-30 Bus system to verify the proposed model in this paper. Experiment results show that bounded rationality affects strategy selection and reduces attacker's payoffs.

With the increasingly fierce competition for marine resources, underwater acoustic communication as the main form of underwater communication, its security has received more and more attention. The traditional underwater acoustic communication technology with fixed frequency and modulation may cause information leakage and location exposure of the communicating platform. The communication based on ocean noise is helpful to improve the communication security. This article proposes a new covert underwater acoustic communication scheme based on ship-radiated noise and chaos signal. Firstly, the characteristics of chaos signal is studied, which is generated by the Chebyshev sequence, and the reconstruction process of ship-radiated noise is analyzed. Secondly, we generate mimic ship-radiated noise containing the secret message based on the time and frequency feature of the chaos signal and ship-radiated noise. Finally, resampling technology is used to equalize the Doppler frequency shift, and time inversion mirror technology is used to perform multipath channel equalization. Simulation and sea experiment verify the effectiveness and feasibility of this scheme. In the sea experiments, the proposed scheme can reliably transmit over 40 per second when the communication distance is about 10km.

It is challenging to detect emotions on user-generated contents (UGC) because it tends to have sparse emotional semantics, multiple emotions in the same text, and a fast update of emotional expression. Word embedding can extract high-level features in words to enrich their semantics. The convolutional neural networks (CNNs) can make model training more efficient to adapt to fast-changing emotional expressions, but its original pooling operation can not simultaneously filter out multiple emotional features that are beneficial to classification results, which is not suitable for UGC. In this paper, we propose a self-attentive convolutional neural networks (SACNNs) trained on top of pre-trained word vectors for emotion detection on UGC, which reserves different kinds of emotion information, avoids the loss of emotional aspects in the pooling process of CNNs structure, and increases the interpretability of the model by visualizing the extraction process of features. Meanwhile, the convergence speed of the model training is accelerated, and the model is updated in real time to detect emotions with rich and novel expressions. The proposed model combines the CNNs and the self-attention mechanism, the self-attention mechanism can select key emotional features after convolution. We evaluate the proposed model with two datasets from NLPCC 2014 and SemEval 2018 Task 1. The experimental results show that our model obtain significant performance than baselines in multi-label classification on UGC. In addition, the experimental results and the rationality of the self-attention mechanism are analyzed in detail, and the influential convolutional filter windows are visualized based on attention weights.

For updating the tracking models, most existing approaches have an assumption that the target changes smoothly over time. Despite their success in some cases, these approaches struggle in dealing with occlusion, illumination changes and abrupt motion which may break the temporal smoothness assumption. To tackle this problem, in this paper we propose a tree-guided visual tracking model based on the multimodality correlation filter which could estimate the target state according to the most reliable information in previous frames. We maintain a representative target state set in a tree model over the whole tracking process. Ideally, the tree model is able to capture all the landmark states of the target, and provides a confident template for the correlation filter. Therefore, we propose an optimal updating strategy to record the most recent stable and representative states for tree updating. By utilizing stable target-states for template training, the multi-modality correlation filter is able to output a more accurate target position than the baseline and the SOTA (state-of-the-art) methods. Tested on the OTB50 (object tracking benchmark) and OTB100 dataset, the proposed TGMCF has demonstrated outstanding performance on several typical tracking difficulties and overall comparative results with the SOTA trackers are obtained on several public tracking benchmarks.

Learning the correlation among labels is a standing-problem in the multi-label image recognition task. The label correlation is the key to solve the multi-label classification but it is too abstract to model. Most solutions try to learn image label dependencies to improve multi-label classification performance. However, they have ignored two more realistic problems: object scale inconsistent and label tail (category imbalance). These two problems will impact the bad influence on the classification model. To tackle these two problems and learn the label correlations, we propose feature attention network (FAN) which contains feature refinement network and correlation learning network. FAN builds top-down feature fusion mechanism to refine more important features and learn the correlations among convolutional features from FAN to indirect learn the label dependencies. Following our proposed solution, we achieve performed classification accuracy on MSCOCO 2014 and VOC 2007 dataset.

Rainstorms, insect swarms, and galloping horses produce “sound textures,”which are the resulting natural sounds of many similar acoustic events. With new achievements emerging regularly for generative models, the deep convolutional neural network (CNN) has proven to be a tremendously successful approach for image and sound synthesis. Existing state-of-the-art sound texture generative models simply treat sound texture signals as 1-D images while discarding the difference between the human vision and auditory systems. This paper considers mel-frequency statistical features, which are designed according to the human auditory system and have been viewed as the dominant features for sound identification. We first construct a CNN structure for extracting mel-frequency features from sounds losslessly. This structure is called mel-frequency CNN (MF-CNN). Next, we investigate a novel sound texture generative model by incorporating the MF-CNN into a convolutional generative network composed of cascading upsampling groups. A jointly alternating back propagation algorithm is proposed to train the overall network. The feedback of the MF-CNN is used to advise the gradients in the inferential and learning back propagation to make the mel-frequency features of the synthesized sounds more similar to the natural ones. Moreover, the proposed generative model can also be extended to other sound synthesis tasks.