Search results

The current popular image processing technologies based on convolutional neural network have the characteristics of large computation, high storage cost and low accuracy for tiny defect detection, which is contrary to the high real-time and accuracy, limited computing resources and storage required by industrial applications. Therefore, an improved YOLOv4 named as YOLOv4-Defect is proposed aim to solve the above problems.

On the one hand, this study performs multi-dimensional compression processing on the feature extraction network of YOLOv4 to simplify the model and improve the feature extraction ability of the model through knowledge distillation. On the other hand, a prediction scale with more detailed receptive field is added to optimize the model structure, which can improve the detection performance for tiny defects.

The effectiveness of the method is verified by public data sets NEU-CLS and DAGM 2007, and the steel ingot data set collected in the actual industrial field. The experimental results demonstrated that the proposed YOLOv4-Defect method can greatly improve the recognition efficiency and accuracy and reduce the size and computation consumption of the model.

This paper proposed an improved YOLOv4 named as YOLOv4-Defect for the detection of surface defect, which is conducive to application in various industrial scenarios with limited storage and computing resources, and meets the requirements of high real-time and precision.

With the advent of technology, a huge amount of data is being transmitted and received through the internet. Large bandwidth and storage are required for the exchange of data and storage, respectively. Hence, compression of the data which is to be transmitted over the channel is unavoidable. The main purpose of the proposed system is to use the bandwidth effectively. The videos are compressed at the transmitter’s end and reconstructed at the receiver’s end. Compression techniques even help for smaller storage requirements.

The paper proposes a novel compression technique for three-dimensional (3D) videos using a zig-zag 3D discrete cosine transform. The method operates a 3D discrete cosine transform on the videos, followed by a zig-zag scanning process. Finally, to convert the data into a single bit stream for transmission, a run-length encoding technique is used. The videos are reconstructed by using the inverse 3D discrete cosine transform, inverse zig-zag scanning (quantization) and inverse run length coding techniques. The proposed method is simple and reduces the complexity of the convolutional techniques.

Coding reduction, code word reduction, peak signal to noise ratio (PSNR), mean square error, compression percent and compression ratio values are calculated, and the dominance of the proposed method over the convolutional methods is seen.

With zig-zag quantization and run length encoding using 3D discrete cosine transform for 3D video compression, gives compression up to 90% with a PSNR of 41.98 dB. The proposed method can be used in multimedia applications where bandwidth, storage and data expenses are the major issues.

The aim of this study is to assess the opportunity for the development of hydrocarbon transportation using high-strength steel (HSS) in pipeline construction in terms of cost savings and reliability.

Several optimizations of pipeline design and operations were performed to estimate the total life-cycle cost variation associated with different grades of high-strength steel. The generalized reduced gradient (GRG) method was used in an Excel table to determine optimal total life cycle each pipeline. Variables used in this optimization with respect to each steel grade were as follows: pipeline external diameter, wall thickness, number of compression stations and installed power in each compression station. The reliability of a pipeline with optimal cost was assessed to highlight the impact of steel grade on pipeline reliability.

The study showed that the cost reduction is strongly dependent on the adopted gas pipeline configuration. The number of compression stations and external diameter are the main factors influencing the pipeline total life cycle cost, while the steel price seems to have a minor effect, the reduction of the gas pipeline total life cycle does not exceed 5% even with a 50% difference in pipe steel prices between X70 and X100 steels. On the other side, for the same external diameter, X100 steel presents better pipeline reliability against carbonic corrosion compared to X70 steel.

The main contribution of this study is to provide a decision-support tool to help pipeline constructors enhance the profitability of natural gas transmission pipelines. The optimization method used is simple to use for design engineers during a feasibility study.

The present study presents one step to fill the gap concerning the question of balancing the trade-off between cost savings and structural reliability in high-strength steel pipelines during the early stages of feasibility studies. The optimal design and operations parameters ensuring cost savings on total life cycle costs are identified via an optimization method. The impact of selected optimal parameters on the long-term pipeline service life was estimated via a structural reliability analysis.

Minimizing the energy consumption in a wireless sensor node is important for lengthening the lifetime of a battery. Radio transmission is the most energy-consuming task in a wireless sensor node, and by compressing the sensor data in the online mode, it is possible to reduce the number of transmission periods. This study aims to demonstrate that temporal compression methods present an effective method for lengthening the lifetime of a battery-powered wireless sensor node.

In this study, the energy consumption of LoRa-based sensor node was evaluated and measured. The experiments were conducted with different LoRaWAN data rate parameters, with and without compression algorithms implemented to compress sensor data in the online mode. The effect of temporal compression algorithms on the overall energy consumption was measured.

Energy consumption was measured with different LoRaWAN spreading factors. The LoRaWAN transmission energy consumption significantly depends on the spreading factor used. The other significant factors affecting the LoRa-based sensor node energy consumption are the measurement interval and sleep mode current consumption. The results show that temporal compression algorithms are an effective method for reducing the energy consumption of a LoRa sensor node by reducing the number of LoRa transmission periods.

This paper presents with a practical case that it is possible to reduce the overall energy consumption of a wireless sensor node by compressing sensor data in online mode with simple temporal compression algorithms.

The trend of “Deep Learning for Internet of Things (IoT)” has gained fresh momentum with enormous upcoming applications employing these models as their processing engine and Cloud as their resource giant. But this picture leads to underutilization of ever-increasing device pool of IoT that has already passed 15 billion mark in 2015. Thus, it is high time to explore a different approach to tackle this issue, keeping in view the characteristics and needs of the two fields. Processing at the Edge can boost applications with real-time deadlines while complementing security.

This review paper contributes towards three cardinal directions of research in the field of DL for IoT. The first section covers the categories of IoT devices and how Fog can aid in overcoming the underutilization of millions of devices, forming the realm of the things for IoT. The second direction handles the issue of immense computational requirements of DL models by uncovering specific compression techniques. An appropriate combination of these techniques, including regularization, quantization, and pruning, can aid in building an effective compression pipeline for establishing DL models for IoT use-cases. The third direction incorporates both these views and introduces a novel approach of parallelization for setting up a distributed systems view of DL for IoT.

DL models are growing deeper with every passing year. Well-coordinated distributed execution of such models using Fog displays a promising future for the IoT application realm. It is realized that a vertically partitioned compressed deep model can handle the trade-off between size, accuracy, communication overhead, bandwidth utilization, and latency but at the expense of an additionally considerable memory footprint. To reduce the memory budget, we propose to exploit Hashed Nets as potentially favorable candidates for distributed frameworks. However, the critical point between accuracy and size for such models needs further investigation.

To the best of our knowledge, no study has explored the inherent parallelism in deep neural network architectures for their efficient distribution over the Edge-Fog continuum. Besides covering techniques and frameworks that have tried to bring inference to the Edge, the review uncovers significant issues and possible future directions for endorsing deep models as processing engines for real-time IoT. The study is directed to both researchers and industrialists to take on various applications to the Edge for better user experience.

The development of a theoretical model for predicting the combustion, performance and emission characteristics of a cylinder head porous medium engine becomes necessary due to imposed requirements from the viewpoint of power, efficiency and toxic gases in the exhaust. The cylinder head porous medium engine was found to have superior combustion, performance and emission characteristics when compared to a conventional diesel engine. The paper aims to discuss these issues.

Due to heterogeneous and transient operation of diesel engine under conventional and porous medium mode, the combustion process becomes complex, and achieving a pure analytical solution to the problem was difficult. Although, closer accuracy of correlation between the computer models and the experimental results is improbable, the computer model will give an opportunity to quantify the combustion and heat transfer processes and thus the performance and emission characteristics of an engine.

In this research work, a theoretical model was developed to predict the combustion, performance and emission characteristics of a cylinder head porous medium engine through two-zone combustion modeling technique, and the results were validated through experimentation.

The two-zone model developed by using programming language C for the purpose of predicting combustion, performance and emission characteristics of a porous medium engine is the first of its kind.

Fabrics’ thermal properties greatly influence human comfort during wear. For this reason, fabrics with optimum thermal properties need to be developed. This paper aims to investigate the effect of weft yarn twist levels on thermal and surface properties of 100 per cent cotton woven fabrics.

Five types of plain woven cotton fabrics were manufactured using weft yarns with 900, 905, 910, 915 and 920 twists/meter (Tpm). The other parameters of the samples as count, thread density and fabric structures were kept constant. Fabric thermal properties were evaluated by measuring its thermal conductivity, thermal resistance, actual insulation, water permeability, air permeability and wicking ability. The fabric compression and surface properties were also evaluated because they contribute to the overall clothing comfort.

The results showed that actual insulation and thermal resistance property decreased with an increase in twists/meter of the weft yarn. However, thermal conductivity does not significantly change while fabric compression reduced with an increase in twist as the surface roughness increased.

Comfort is a fundamental requirement in human daily existence, and it is greatly influenced by clothing, which comes in close contact with the human skin. Fabrics’ thermal properties greatly influence human comfort during wear. For this reason, fabrics with optimum thermal properties need to be developed.

The purpose of this study is to present the state of the art for fiber Bragg grating (FBG) acceleration sensing technologies from two aspects: the principle of the measurement dimension and the principle of the sensing configuration. Some commercial sensors have also been introduced and future work in this field has also been discussed. This paper could provide an important reference for the research community.

This review is to present the state of the art for FBG acceleration sensing technologies from two aspects: the principle of the measurement dimension (one-dimension and multi-dimension) and the principle of the sensing configuration (beam type, radial vibration type, axial vibration type and other composite structures).

The current research on developing FBG acceleration sensors is mainly focused on the sensing method, the construction and design of the elastic structure and the design of a new information detection method. This paper hypothesizes that in the future, the following research trends will be strengthened: common single-mode fiber grating of the low cost and high utilization rate; high sensitivity and strength special fiber grating; multi-core fiber grating for measuring single-parameter multi-dimensional information or multi-parameter information; demodulating equipment of low cost, small volume and high sampling frequency.

The principle of the measurement dimension and principle of the sensing configuration for FBG acceleration sensors have been introduced, which could provide an important reference for the research community.

The developments of complex systems have increased the demand for condition monitoring techniques so as to maximize operational availability and safety while decreasing the costs. Signal analysis is one of the methods used to develop condition monitoring in order to extract important information contained in the sensory signals, which can be used for health assessment. However, extraction of such information from collected data in a practical working environment is always a great challenge as sensory signals are usually multi-dimensional and obscured by noise. The paper aims to discuss this issue.

This paper presents a method for trends extraction from multi-dimensional sensory data, which are then used for machinery health monitoring and maintenance needs. The proposed method is based on extracting successive features from machinery sensory signals. Then, unsupervised feature selection on the features domain is applied without making any assumptions concerning the source of the signals and the number of the extracted features. Finally, empirical mode decomposition (EMD) algorithm is applied on the projected features with the purpose of following the evolution of data in a compact representation over time.

The method is demonstrated on accelerated degradation data set of bearings acquired from PRONOSTIA experimental platform and a second data set acquired form NASA repository.

The method showed that it is able to extract interesting signal trends which can be used for health monitoring and remaining useful life prediction.

The purpose of this paper is to develop a theoretical framework to explain how leaders emerge in teams that lack a hierarchical structure. This framework emphasizes the perceptual processes through which team members determine whether or not an individual fits with the task, the group, and the situational context.

This paper builds on prior leadership research to develop a theoretical framework of emergent leadership, a testable model, and research propositions.

The authors suggest that team members’ perceptions of leadership fit depend on the potential leader's domain competence, fluid intelligence, willingness to serve, credibility, and goal attainment. A conceptual framework is developed to suggest these attributes combine to create perceptions of leadership fit that must correspond to the degree of stress in the situational context, which varies according to task criticality and time compression. The framework suggests that an individual perceived by team members to exhibit characteristics that fit with the situation will likely emerge as the leader.

This paper focusses on emergent leadership, but does not address which path to leadership may be best. Future research may also address group dynamics (i.e. cohesion or group potency) and the implications for leader emergence.

This research contributes to the discipline by suggesting a potential path of leader emergence in multiple contexts of situational stress and leader behaviors.