Search results

The purpose of this paper is to design a robust tracking algorithm which is suitable for the real-time requirement and solves the mistake labeling issue in the appearance model of trackers with the spare features.

This paper proposes a tracker to select the most discriminative randomly projected ferns and integrates a coarse-to-fine search strategy in this framework. First, the authors exploit multiple instance boosting learning to maximize the bag likelihood and select randomly projected fern from feature pool to degrade the effect of mistake labeling. Second, a coarse-to-fine search approach is first integrated into the framework of multiple instance learning (MIL) for less detections.

The quantitative and qualitative experiments demonstrate that the tracker has shown favorable performance in efficiency and effective among the competitors of tracking algorithms.

The proposed method selects the feature from the compressive domain by MIL AnyBoost and integrates the coarse-to-fine search strategy first to reduce the burden of detection. This paper designs a tracker with high speed and favorable results which is more suitable for real-time scene.

This study aims to present a conceptual framework aimed at promoting educational innovation in supply chain management and logistics (SCM&L). The framework can help to design active learning experiences regarding student learning outcomes that tackle current challenges in the discipline. Emphasizing the significance of linking students’ learning to real-world scenarios, the framework enables reflective learning through hands-on engagement in a constructive alignment, overcoming existing pedagogical limitations in the field.

This study presents a qualitative research methodology that relies on the case study method. Three instances are presented to illustrate educational efforts of active learning in countries of Latin America, Bolivia, Mexico and Peru, linking real-world relevant situations to disciplinary teaching and learning.

The innovative learning experiences introduced in this study transform real-world SCM&L operations into distinctive educational opportunities. These experiences facilitate learning not only within traditional classrooms but also in urban areas of the Latin American region, enabling students to interact with educational partners in authentic settings to achieve their intended learning outcomes. These experiences are characterized by their focus on establishing meaningful connections between learning and local communities, businesses or specific contexts.

The study recognizes various limitations of conceptual, methodological, execution-related and research process aspects. First, not all academics in the SCM&L discipline may universally acknowledge the importance of educational innovation and active learning experiences because of limited pedagogical awareness. Moreover, execution-related limitations arise from the demanding nature of incorporating active pedagogical approaches into courses, as they can be resource-intensive and time-consuming. Regarding research process limitations, the case study limits generalizability and broader inferences because of its particular views and locations, which require further investigation with other instances across other disciplines and geographical regions for validation.

The practical implementation of this framework within the MIT SCALE network for Latin America and the Caribbean (LAC) demonstrates its potential in meeting diverse academic and institutional expectations and providing educational benefits to students.

The study makes a valuable contribution to prioritizing and coordinating pedagogical research by investigating the success of learning outcomes achieved through active and experiential implementations in various contexts. It provides inspiring examples of innovative learning experiences that can drive new developments not only within the LAC region but also in other areas, prompting a shift away from traditional educational approaches.

This research presents a conceptual framework, which is developed from the insights obtained in the three learning experiences to guide future efforts in SCM&L education. The findings demonstrate how to structure active learning experiences based on authentic assessment and illustrate the potential for increased cooperation among institutions in Latin America. It also promotes the recognition of novel SCM&L active learning experiences and highlights some of the benefits of this approach.

The paper aims to address the tracking algorithm based on deep learning and four deep learning tracking models developed. They compared with each other to prevent collision and to obtain target tracking in autonomous aircraft.

First, to follow the visual target, the detection methods were used and then the tracking methods were examined. Here, four models (deep convolutional neural networks (DCNN), deep convolutional neural networks with fine-tuning (DCNNFN), transfer learning with deep convolutional neural network (TLDCNN) and fine-tuning deep convolutional neural network with transfer learning (FNDCNNTL)) were developed.

The training time of DCNN took 9 min 33 s, while the accuracy percentage was calculated as 84%. In DCNNFN, the training time of the network was calculated as 4 min 26 s and the accuracy percentage was 91%. The training of TLDCNN) took 34 min and 49 s and the accuracy percentage was calculated as 95%. With FNDCNNTL, the training time of the network was calculated as 34 min 33 s and the accuracy percentage was nearly 100%.

Compared to the results in the literature ranging from 89.4% to 95.6%, using FNDCNNTL, better results were found in the paper.

In the community of visual tracking or object tracking, discriminatively learned correlation filter (DCF) has gained more importance. When it comes to speed, DCF gives the best performance. The main objective of this study is to anticipate the object visually. For tracking the object visually, the authors proposed a new model based on the convolutional regression technique. Features like HOG & Harris are used for the process of feature extraction. The proposed method will give the best results when compared to other existing methods.

This paper introduces the concept and research status of tracks; later the authors focus on the representative applications of deep learning in visual tracking.

Better tracking algorithms are not mentioned in the existing method.

Visual tracking is the ability to control eye movements using the oculomotor system (vision and eye muscles working together). Visual tracking plays an important role when it comes to identifying an object and matching it with the database images. In visual tracking, deep learning has achieved great success.

The authors implement the multiple tracking methods, for better tracking purpose.

The main theme of this paper is to review the state-of-the-art tracking methods depending on deep learning. First, we introduce the visual tracking that is carried out manually, and secondly, we studied different existing methods of visual tracking based on deep learning. For every paper, we explained the analysis and drawbacks of that tracking method. This paper introduces the concept and research status of tracks, later we focus on the representative applications of deep learning in visual tracking.

Automatic defect detection is a fundamental and vital topic in the research field of industrial intelligence. In this work, the authors develop a more flexible deep learning method for the industrial defect detection.

The authors propose a unified framework for detecting defects in industrial products or planar surfaces based on an end-to-end learning strategy. A lightweight deep learning architecture for blade defect detection is specifically demonstrated. In addition, a blade defect data set is collected with the dual-arm image collection system.

Numerous experiments are conducted on the collected data set, and experimental results demonstrate that the proposed system can achieve satisfactory performance over other methods. Furthermore, the data equalization operation helps for a better defect detection result.

An end-to-end learning framework is established for defect detection. Although the adopted fully convolutional network has been extensively used for semantic segmentation in images, to the best knowledge of the authors, it has not been used for industrial defect detection. To remedy the difficulties of blade defect detection which has been analyzed above, the authors develop a new network architecture which integrates the residue learning to perform the efficient defect detection. A dual-arm data collection platform is constructed and extensive experimental validation are conducted.

Stock market forecasters are focusing to create a positive approach for predicting the stock price. The fundamental principle of an effective stock market prediction is not only to produce the maximum outcomes but also to reduce the unreliable stock price estimate. In the stock market, sentiment analysis enables people for making educated decisions regarding the investment in a business. Moreover, the stock analysis identifies the business of an organization or a company. In fact, the prediction of stock prices is more complex due to high volatile nature that varies a large range of investor sentiment, economic and political factors, changes in leadership and other factors. This prediction often becomes ineffective, while considering only the historical data or textural information. Attempts are made to make the prediction more precise with the news sentiment along with the stock price information.

This paper introduces a prediction framework via sentiment analysis. Thereby, the stock data and news sentiment data are also considered. From the stock data, technical indicator-based features like moving average convergence divergence (MACD), relative strength index (RSI) and moving average (MA) are extracted. At the same time, the news data are processed to determine the sentiments by certain processes like (1) pre-processing, where keyword extraction and sentiment categorization process takes place; (2) keyword extraction, where WordNet and sentiment categorization process is done; (3) feature extraction, where Proposed holoentropy based features is extracted. (4) Classification, deep neural network is used that returns the sentiment output. To make the system more accurate on predicting the sentiment, the training of NN is carried out by self-improved whale optimization algorithm (SIWOA). Finally, optimized deep belief network (DBN) is used to predict the stock that considers the features of stock data and sentiment results from news data. Here, the weights of DBN are tuned by the new SIWOA.

The performance of the adopted scheme is computed over the existing models in terms of certain measures. The stock dataset includes two companies such as Reliance Communications and Relaxo Footwear. In addition, each company consists of three datasets (a) in daily option, set start day 1-1-2019 and end day 1-12-2020, (b) in monthly option, set start Jan 2000 and end Dec 2020 and (c) in yearly option, set year 2000. Moreover, the adopted NN + DBN + SIWOA model was computed over the traditional classifiers like LSTM, NN + RF, NN + MLP and NN + SVM; also, it was compared over the existing optimization algorithms like NN + DBN + MFO, NN + DBN + CSA, NN + DBN + WOA and NN + DBN + PSO, correspondingly. Further, the performance was calculated based on the learning percentage that ranges from 60, 70, 80 and 90 in terms of certain measures like MAE, MSE and RMSE for six datasets. On observing the graph, the MAE of the adopted NN + DBN + SIWOA model was 91.67, 80, 91.11 and 93.33% superior to the existing classifiers like LSTM, NN + RF, NN + MLP and NN + SVM, respectively for dataset 1. The proposed NN + DBN + SIWOA method holds minimum MAE value of (∼0.21) at learning percentage 80 for dataset 1; whereas, the traditional models holds the value for NN + DBN + CSA (∼1.20), NN + DBN + MFO (∼1.21), NN + DBN + PSO (∼0.23) and NN + DBN + WOA (∼0.25), respectively. From the table, it was clear that the RMSRE of the proposed NN + DBN + SIWOA model was 3.14, 1.08, 1.38 and 15.28% better than the existing classifiers like LSTM, NN + RF, NN + MLP and NN + SVM, respectively, for dataset 6. In addition, he MSE of the adopted NN + DBN + SIWOA method attain lower values (∼54944.41) for dataset 2 than other existing schemes like NN + DBN + CSA(∼9.43), NN + DBN + MFO (∼56728.68), NN + DBN + PSO (∼2.95) and NN + DBN + WOA (∼56767.88), respectively.

This paper has introduced a prediction framework via sentiment analysis. Thereby, along with the stock data and news sentiment data were also considered. From the stock data, technical indicator based features like MACD, RSI and MA are extracted. Therefore, the proposed work was said to be much appropriate for stock market prediction.

Monitoring corporate credit risk (CCR) has traditionally relied on such indicators as income, debt and inventory at a company level. These data are usually released on a quarterly or annual basis by the target company and include, exclusively, the financial data of the target company. As a result of this exclusiveness, the models for monitoring credit risk usually fail to account for some significant information from different sources or channels, like the data of its supply chain partner companies and other closely relevant data yet available from public networks, and it is these seldom used data that can help unveil the immediate CCR changes and how the risk is being propagated along the supply chain. This study aims to discuss the a forementioned issues.

Going beyond the existing CCR prediction data, this study intends to address the impact of supply chain data and network activity data on CCR prediction, by integrating machine learning technology into the prediction to verify whether adding new data can improve the predictability.

The results show that the predictive errors of the datasets after adding supply chain data and network activity data to them are made the ever least. Moreover, intelligent algorithms like support vector machine (SVM), compared to traditionally used methods, are better at processing nonlinear datasets and mining complex relationships between multi-variable indicators for CCR evaluation.

This study indicates that bringing in more information of multiple data sources combined with intelligent algorithms can help companies prevent risk spillovers in the supply chain from causing harm to the company, and, as well, help customers evaluate the creditworthiness of the entity to lessen the risk of their investment.

This paper aims to quantify the quality of peer reviews, evaluate them from different perspectives and develop a model to predict the review quality. In addition, this paper investigates effective features to distinguish the reviews' quality. 

First, a fine-grained data set including peer review data, citations and review conformity scores was constructed. Second, metrics were proposed to evaluate the quality of peer reviews from three aspects. Third, five categories of features were proposed in terms of reviews, submissions and responses using natural language processing (NLP) techniques. Finally, different machine learning models were applied to predict the review quality, and feature analysis was performed to understand effective features.

The analysis results revealed that reviewers become more conservative and the review quality becomes worse over time in terms of these indicators. Among the three models, random forest model achieves the best performance on all three tasks. Sentiment polarity, review length, response length and readability are important factors that distinguish peer reviews’ quality, which can help meta-reviewers value more worthy reviews when making final decisions.

This study provides a new perspective for assessing review quality. Another originality of the research lies in the proposal of a novelty task that predict review quality. To address this task, a novel model was proposed which incorporated various of feature sets, thereby deepening the understanding of peer reviews.

The present research work is carried out for determining haemoprotozoan diseases in cattle and breast cancer diseases in humans at early stage. The combination of LeNet and bidirectional long short-term memory (Bi-LSTM) model is used for the classification of heamoprotazoan samples into three classes such as theileriosis, babesiosis and anaplasmosis. Also, BreaKHis dataset image samples are classified into two major classes as malignant and benign. The hyperparameter optimization is used for selecting the prominent features. The main objective of this approach is to overcome the manual identification and classification of samples into different haemoprotozoan diseases in cattle. The traditional laboratory approach of identification is time-consuming and requires human expertise. The proposed methodology will help to identify and classify the heamoprotozoan disease in early stage without much of human involvement.

LeNet-based Bi-LSTM model is used for the classification of pathology images into babesiosis, anaplasmosis, theileriosis and breast images classified into malignant or benign. An optimization-based super pixel clustering algorithm is used for segmentation once the normalization of histopathology images is conducted. The edge information in the normalized images is considered for identifying the irregular shape regions of images, which are structurally meaningful. Also, it is compared with another segmentation approach circular Hough Transform (CHT). The CHT is used to separate the nuclei from non-nuclei. The Canny edge detection and gaussian filter is used for extracting the edges before sending to CHT.

The existing methods such as artificial neural network (ANN), convolution neural network (CNN), recurrent neural network (RNN), LSTM and Bi-LSTM model have been compared with the proposed hyperparameter optimization approach with LeNET and Bi-LSTM. The results obtained by the proposed hyperparameter optimization-Bi-LSTM model showed the accuracy of 98.99% when compared to existing models like Ensemble of Deep Learning Models of 95.29% and Modified ReliefF Algorithm of 95.94%.

In contrast to earlier research done using Modified ReliefF, the suggested LeNet with Bi-LSTM model, there is an improvement in accuracy, precision and F-score significantly. The real time data set is used for the heamoprotozoan disease samples. Also, for anaplasmosis and babesiosis, the second set of datasets were used which are coloured datasets obtained by adding a chemical acetone and stain.

Diabetic retinopathy (DR) is a central root of blindness all over the world. Though DR is tough to diagnose in starting stages, and the detection procedure might be time-consuming even for qualified experts. Nowadays, intelligent disease detection techniques are extremely acceptable for progress analysis and recognition of various diseases. Therefore, a computer-aided diagnosis scheme based on intelligent learning approaches is intended to propose for diagnosing DR effectively using a benchmark dataset.

The proposed DR diagnostic procedure involves four main steps: (1) image pre-processing, (2) blood vessel segmentation, (3) feature extraction, and (4) classification. Initially, the retinal fundus image is taken for pre-processing with the help of Contrast Limited Adaptive Histogram Equalization (CLAHE) and average filter. In the next step, the blood vessel segmentation is carried out using a segmentation process with optimized gray-level thresholding. Once the blood vessels are extracted, feature extraction is done, using Local Binary Pattern (LBP), Texture Energy Measurement (TEM based on Laws of Texture Energy), and two entropy computations – Shanon's entropy, and Kapur's entropy. These collected features are subjected to a classifier called Neural Network (NN) with an optimized training algorithm. Both the gray-level thresholding and NN is enhanced by the Modified Levy Updated-Dragonfly Algorithm (MLU-DA), which operates to maximize the segmentation accuracy and to reduce the error difference between the predicted and actual outcomes of the NN. Finally, this classification error can correctly prove the efficiency of the proposed DR detection model.

The overall accuracy of the proposed MLU-DA was 16.6% superior to conventional classifiers, and the precision of the developed MLU-DA was 22% better than LM-NN, 16.6% better than PSO-NN, GWO-NN, and DA-NN. Finally, it is concluded that the implemented MLU-DA outperformed state-of-the-art algorithms in detecting DR.

This paper adopts the latest optimization algorithm called MLU-DA-Neural Network with optimal gray-level thresholding for detecting diabetic retinopathy disease. This is the first work utilizes MLU-DA-based Neural Network for computer-aided Diabetic Retinopathy diagnosis.