Search results

This study aims to clarify influencing factors of overcapacity of new energy enterprises in China and accurately predict whether these enterprises have overcapacity.

Based on relevant data including the experience and evidence from the capital market in China, the research establishes a generic univariate selection-comparative machine learning model to study relevant factors that affect overcapacity of new energy enterprises from five dimensions. These include the governmental intervention, market demand, corporate finance, corporate governance and corporate decision. Moreover, the bridging approach is used to strengthen findings from quantitative studies via the results from qualitative studies.

The authors' results show that the overcapacity of new energy enterprises in China is brought out by the combined effect of governmental intervention corporate governance and corporate decision. Governmental interventions increase the overcapacity risk of new energy enterprises mainly by distorting investment behaviors of enterprises. Corporate decision and corporate governance factors affect the overcapacity mainly by regulating the degree of overconfidence of the management team and the agency cost. Among the eight comparable integrated models, generic univariate selection-bagging exhibits the optimal comprehensive generalization performance and its area under the receiver operating characteristic curve Area under curve (AUC) accuracy precision and recall are 0.719, 0.960, 0.975 and 0.983, respectively.

The proposed integrated model analyzes causes and predicts presence of overcapacity of new energy enterprises to help governments to formulate appropriate strategies to deal with overcapacity and new energy enterprises to optimize resource allocation. Ten main features which affect the overcapacity of new energy enterprises in China are identified through generic univariate selection model. Through the bridging approach, the impact of the main features on the overcapacity of new energy enterprises and the mechanism of the influence are analyzed.

The purpose of this paper is to propose a whole set of methods for underwater target detection, because most underwater objects have small samples, low quality underwater images problems such as detail loss, low contrast and color distortion, and verify the feasibility of the proposed methods through experiments.

The improved RGHS algorithm to enhance the original underwater target image is proposed, and then the YOLOv4 deep learning network for underwater small sample targets detection is improved based on the combination of traditional data expansion method and Mosaic algorithm, expanding the feature extraction capability with SPP (Spatial Pyramid Pooling) module after each feature extraction layer to extract richer feature information.

The experimental results, using the official dataset, reveal a 3.5% increase in average detection accuracy for three types of underwater biological targets compared to the traditional YOLOv4 algorithm. In underwater robot application testing, the proposed method achieves an impressive 94.73% average detection accuracy for the three types of underwater biological targets.

Underwater target detection is an important task for underwater robot application. However, most underwater targets have the characteristics of small samples, and the detection of small sample targets is a comprehensive problem because it is affected by the quality of underwater images. This paper provides a whole set of methods to solve the problems, which is of great significance to the application of underwater robot.

Single-shot multi-category clothing recognition and retrieval play a crucial role in online searching and offline settlement scenarios. Existing clothing recognition methods based on RGBD clothing images often suffer from high-dimensional feature representations, leading to compromised performance and efficiency.

To address this issue, this paper proposes a novel method called Manifold Embedded Discriminative Feature Selection (MEDFS) to select global and local features, thereby reducing the dimensionality of the feature representation and improving performance. Specifically, by combining three global features and three local features, a low-dimensional embedding is constructed to capture the correlations between features and categories. The MEDFS method designs an optimization framework utilizing manifold mapping and sparse regularization to achieve feature selection. The optimization objective is solved using an alternating iterative strategy, ensuring convergence.

Empirical studies conducted on a publicly available RGBD clothing image dataset demonstrate that the proposed MEDFS method achieves highly competitive clothing classification performance while maintaining efficiency in clothing recognition and retrieval.

This paper introduces a novel approach for multi-category clothing recognition and retrieval, incorporating the selection of global and local features. The proposed method holds potential for practical applications in real-world clothing scenarios.

Several chaotic system-based encryption techniques have been presented in recent years to protect digital images using cryptography. The challenges of key distribution and administration make symmetric encryption difficult. The purpose of this paper is to address these concerns, the novel hybrid partial differential elliptical Rubik’s cube algorithm is developed in this study as an asymmetric image encryption approach. This novel algorithm generates a random weighted matrix, and uses the masking method on image pixels with Rubik’s cube principle. Security analysis has been conducted, it enhances and increases the reliability of the proposed algorithm against a variety of attacks including statistical and differential attacks.

In this light, a differential elliptical model is designed with two phases for image encryption and decryption. A modified image is achieved by rotating and mixing intensities of rows and columns with a masking matrix derived from the key generation technique using a unique approach based on the elliptic curve and Rubik’s cube principle.

To evaluate the security level, the proposed algorithm is tested with statistical and differential attacks on a different set of test images with peak signal-to-noise ratio, unified average changed intensity and number of pixel change rate performance metrics. These results proved that the proposed image encryption method is completely reliable and enhances image security during transmission.

The elliptic curve–based encryption is hard to break by hackers and adding a Rubik’s cube principle makes it even more complex and nearly impossible to decode. The proposed method provides reduced key size.

Image segmentation is one of the most essential tasks in image processing applications. It is a valuable tool in many oriented applications such as health-care systems, pattern recognition, traffic control, surveillance systems, etc. However, an accurate segmentation is a critical task since finding a correct model that fits a different type of image processing application is a persistent problem. This paper develops a novel segmentation model that aims to be a unified model using any kind of image processing application. The proposed precise and parallel segmentation model (PPSM) combines the three benchmark distribution thresholding techniques to estimate an optimum threshold value that leads to optimum extraction of the segmented region: Gaussian, lognormal and gamma distributions. Moreover, a parallel boosting algorithm is proposed to improve the performance of the developed segmentation algorithm and minimize its computational cost. To evaluate the effectiveness of the proposed PPSM, different benchmark data sets for image segmentation are used such as Planet Hunters 2 (PH2), the International Skin Imaging Collaboration (ISIC), Microsoft Research in Cambridge (MSRC), the Berkley Segmentation Benchmark Data set (BSDS) and Common Objects in COntext (COCO). The obtained results indicate the efficacy of the proposed model in achieving high accuracy with significant processing time reduction compared to other segmentation models and using different types and fields of benchmarking data sets.

The proposed PPSM combines the three benchmark distribution thresholding techniques to estimate an optimum threshold value that leads to optimum extraction of the segmented region: Gaussian, lognormal and gamma distributions.

On the basis of the achieved results, it can be observed that the proposed PPSM–minimum cross-entropy thresholding (PPSM–MCET)-based segmentation model is a robust, accurate and highly consistent method with high-performance ability.

A novel hybrid segmentation model is constructed exploiting a combination of Gaussian, gamma and lognormal distributions using MCET. Moreover, and to provide an accurate and high-performance thresholding with minimum computational cost, the proposed PPSM uses a parallel processing method to minimize the computational effort in MCET computing. The proposed model might be used as a valuable tool in many oriented applications such as health-care systems, pattern recognition, traffic control, surveillance systems, etc.

This paper aims to propose a new deep learning technique to detect the type of material to improve automated construction quality monitoring.

A new data augmentation approach that has improved the model robustness against different illumination conditions and overfitting is proposed. This study uses data augmentation at test time and adds outlier samples to training set to prevent over-fitted network training. For data augmentation at test time, five segments are extracted from each sample image and fed to the network. For these images, the network outputting average values is used as the final prediction. Then, the proposed approach is evaluated on multiple deep networks used as material classifiers. The fully connected layers are removed from the end of the networks, and only convolutional layers are retained.

The proposed method is evaluated on recognizing 11 types of building materials which include 1,231 images taken from several construction sites. Each image resolution is 4,000 × 3,000. The images are captured with different illumination and camera positions. Different illumination conditions lead to trained networks that are more robust against various environmental conditions. Using VGG16 model, an accuracy of 97.35% is achieved outperforming existing approaches.

It is believed that the proposed method presents a new and robust tool for detecting and classifying different material types. The automated detection of material will aid to monitor the quality and see whether the right type of material has been used in the project based on contract specifications. In addition, the proposed model can be used as a guideline for performing quality control (QC) in construction projects based on project quality plan. It can also be used as an input for automated progress monitoring because the material type detection will provide a critical input for object detection.

Several studies have been conducted to perform quality management, but there are some issues that need to be addressed. In most previous studies, a very limited number of material types were examined. In addition, although some studies have reported high accuracy to detect material types (Bunrit et al., 2020), their accuracy is dramatically reduced when they are used to detect materials with similar texture and color. In this research, the authors propose a new method to solve the mentioned shortcomings.

Historical bazaars, a huge treasure of Iranian culture, art and economy, are places for social capital development. Un-supervised management in past decades has led to the demolition and change of historical bazaars and negligence of its different aspects. The present research aims to investigate the resilience of historical bazaars preserving their identity and different developments.

The artificial neural network (ANN) has been applied to investigate the resilience of historical bazaars. This model consists of three main networks for evaluating the resilience of historical networks in terms of adaptability, variability and reactivity.

The ANN proposed to evaluate the resilience of historic bazaars based on the mentioned factors is efficient. By calculating mean squared error (MSE), the model accuracy for evaluating adaptability, variability and reactivity were obtained at 7.62e-25, 2.91e-24 and 1.51e-24. The correlation coefficient was obtained at a significance level of 99%. This indicates the considerable effectiveness of the artificial intelligence model in modeling and predicting the qualitative properties of historical bazaars resilience.

This paper clarifies indexes and components of resilience in terms of adaptability, variability and reactivity. Then, the ANN model is obtained with the least error and very high accuracy that predict the resilience of historical bazaars.

Despite large bodies of research related to the impacts of e-commerce on last-mile logistics and sustainability, there has been limited effort to evaluate urban freight using an equity lens. Therefore, this study proposes a modeling framework that enables researchers and planners to estimate the baseline equity performance of a major e-commerce platform and evaluate equity impacts of possible urban freight management strategies. The study also analyzes the sensitivity of various operational decisions to mitigate bias in the analysis.

The model adapts empirical methodologies from activity-based modeling, transport equity evaluation, and residential freight trip generation (RFTG) to estimate person- and household-level delivery demand and cargo van traffic exposure in 41 U.S. Metropolitan Statistical Areas (MSAs).

Evaluating 12 measurements across varying population segments and spatial units, the study finds robust evidence for racial and socio-economic inequities in last-mile delivery for low-income and, especially, populations of color (POC). By the most conservative measurement, POC are exposed to roughly 35% more cargo van traffic than white populations on average, despite ordering less than half as many packages. The study explores the model’s utility by evaluating a simple scenario that finds marginal equity gains for urban freight management strategies that prioritize line-haul efficiency improvements over those improving intra-neighborhood circulations.

Presents a first effort in building a modeling framework for more equitable decision-making in last-mile delivery operations and broader city planning.

The purpose of this research is to provide a theoretical and practical theory and application that provides understanding and means to manage complex infrastructures.

In this research, complexity, nonlinear, noncontinuous effects and aleatoric and data unknowns are bypassed by directly addressing systems' responses. Graph theory, statistics and digital signal processing (DSP) tools are applied within a theoretical framework of the theory of faults (ToF). Motivational complex infrastructure systems (CISs) are difficult to model. Data are often missing or erroneous, changes are not well documented and processes are not well understood. On top of it, under complexity, stalwart analytical tools have limited predictive power. The aleatoric risk, such as rain and risk cascading from interconnected infrastructures, is unpredictable. Mitigation, response and recovery efforts are adversely affected.

The theory and application are presented and demonstrated by a step-by-step development of an application to a municipal drainage system. A database of faults is analyzed to produce system statistics, spatio-temporal morphology, behavior and traits. The gained understanding is compared to the physical system's design and to its modus operandi. Implications for design and maintenance are inferred; DSP tools to manage the system in real time are developed.

Sociological systems are interest driven. Some events are intentionally created and directed to the benefit and detriment of the opposing parties in a project. Those events may be explained and possibly predicted by understanding power plays, not power functions. For those events, sociological game theories provide better explanatory value than mathematical gain theories.

The theory provides a thematic network for modeling and resolving aleatoric uncertainty in engineering and sociological systems. The framework may be elaborated to fields such as energy, healthcare and critical infrastructure.

ToF provides a framework for the modeling and prediction of faults generated by inherent aleatoric uncertainties in social and technological systems. Therefore, the framework and theory lay the basis for automated monitoring and control of aleatoric uncertainties such as mechanical failures and human errors and the development of mitigation systems.

The contribution of this research is in the provision of an explicatory theory and a management paradigm for complex systems. This theory is applicable to a wide variety of fields from facilities and construction project management to maintenance and from academic studies to commercial use.

Diabetic retinopathy (DR) is one of the dangerous complications of diabetes. Its grade level must be tracked to manage its progress and to start the appropriate decision for treatment in time. Effective automated methods for the detection of DR and the classification of its severity stage are necessary to reduce the burden on ophthalmologists and diagnostic contradictions among manual readers.

In this research, convolutional neural network (CNN) was used based on colored retinal fundus images for the detection of DR and classification of its stages. CNN can recognize sophisticated features on the retina and provides an automatic diagnosis. The pre-trained VGG-16 CNN model was applied using a transfer learning (TL) approach to utilize the already learned parameters in the detection.

By conducting different experiments set up with different severity groupings, the achieved results are promising. The best-achieved accuracies for 2-class, 3-class, 4-class and 5-class classifications are 86.5, 80.5, 63.5 and 73.7, respectively.

In this research, VGG-16 was used to detect and classify DR stages using the TL approach. Different combinations of classes were used in the classification of DR severity stages to illustrate the ability of the model to differentiate between the classes and verify the effect of these changes on the performance of the model.