Practical methods of image data preprocessing for enhancing the performance of deep learning based road crack detection

Road crack detection is an important indicator of road detection. In real life, it is very meaningful work to detect road cracks. With the rapid development of science and technology, especially computer science and technology, quite a lot of methods have been applied to crack detection. Traditional detection methods rely on manual identification, which is inefficient and prone to errors. In addition, the commonly used image processing methods are affected by many factors, such as illumination, road stains, etc., so the results are unstable. In the research on pavement crack detection, many research studies mainly focus on the recognition and classification of cracks, lacking the analysis of the specific characteristics of cracks, and the feature values of cracks cannot be measured. Starting from the deep learning method in computer science and technology, this paper proposes a road crack detection technology based on deep learning. It relies on a new deep dictionary learning and encoding network DDLCN, establishes a new activation function MeLU, and adopts a new differentiable calculation method. The technology relies on the traditional Mask-RCNN algorithm and is implemented after improvement. In the comparison of evaluation indicators, the values of recall, precision, and F1-score reflect certain superiority. Experiments show that the proposed method has good implementability and performance in road crack detection and crack feature measurement.

Road cracks are one of the external manifestations of safety hazards in transportation. At present, the detection and segmentation of road cracks is still an intensively researched issue. With the development of image segmentation technology of the convolutional neural network, the identification of road cracks has also ushered in new opportunities. However, the traditional road crack segmentation method has these three problems: 1. It is susceptible to the influence of complex background noise information. 2. Road cracks usually appear in irregular shapes, which increases the difficulty of model segmentation. 3. The cracks appear discontinuous in the segmentation results. Aiming at these problems, a network segmentation model of HC-Unet++ road crack detection is proposed in this paper. In this network model, a deep parallel feature fusion module is first proposed, one which can effectively detect various irregular shape cracks. Secondly, the SEnet attention mechanism is used to eliminate complex backgrounds to correctly extract crack information. Finally, the Blurpool pooling operation is used to replace the original maximum pooling in order to solve the crack discontinuity of the segmentation results. Through the comparison with some advanced network models, it is found that the HC-Unet++ network model is more precise for the segmentation of road cracks. The experimental results show that the method proposed in this paper has achieved 76.32% mIOU, 82.39% mPA, 85.51% mPrecision, 70.26% dice and Hd95 of 5.05 on the self-made 1040 road crack dataset. Compared with the advanced network model, the HC-Unet++ network model has stronger generalization ability and higher segmentation accuracy, which is more suitable for the segmentation detection of road cracks. Therefore, the HC-Unet++ network model proposed in this paper plays an important role in road maintenance and traffic safety.

This study aimed to propose road crack detection method based on infrared image fusion technology. By analyzing the characteristics of road crack images, this method uses a variety of infrared image fusion methods to process different types of images. The use of this method allows the detection of road cracks, which not only reduces the professional requirements for inspectors, but also improves the accuracy of road crack detection. Based on infrared image processing technology, on the basis of in-depth analysis of infrared image features, a road crack detection method is proposed, which can accurately identify the road crack location, direction, length, and other characteristic information. Experiments showed that this method has a good effect, and can meet the requirement of road crack detection.

Road crack detection is the key link of road maintenance. Only when the cracks are detected can they be repaired. Since cracks are not common and manual detection efficiency is low, inspectors prefer to use machine vision methods for crack detection. Aiming at the problems of high noise, low precision and easy loss of image details in traditional road crack detection, this paper improves the classic image segmentation model Unet and applies it to road crack detection. The new model changes Unet to nested structure and integrates attention mechanism on this basis. The nested Unet can better fuse feature maps from different layers through skip connections and retain the details of road crack images effectively. And the attention mechanism is introduced to suppress the noise in irrelevant regions. The improved model has been evaluated on an expanded road crack dataset containing 9,990 images. According to the experimental results, the model can significantly eliminate noise, improve segmentation accuracy, and retain crack details.

Pavement crack detection is a key technology used to judge whether the road is safe or not. Due to the complex and diverse background of cracks, the traditional crack detection algorithm is difficult to accurately detect cracks. In this paper, YOLOv5 algorithm with strong portability is used to detect road cracks. Mosaic data enhancement method is used to enrich the background of detection targets at the input end and improve the detection effect of small targets. The Backbone CSP structure divides the input into two branches, which greatly reduces the computational load while enhancing the learning performance of the entire convolutional neural network. The experimental results show that the model trained on the public dataset achieves 92% detection accuracy on the test set, and the detection time is 0.04s. The model is directly applied to the vehicle and pedestrian detection datasets, and the detection accuracy is improved by 2%, and the detection time is shortened to 0.028s, indicating that the model has good generalization performance. It can be used for crack detection and quality assessment in complex road scenarios.

Road crack detection is an important task for road safety and road maintenance. In the past, people made use of manual detection methods and tried to use computer vision to detect crack. The most prominent feature in recent years is the use of deep learning. However, there is no good deep learning method for road crack detection under noise. This challenge is faced bravely. First, a noise crack dataset is proposed, consisting of multiple noise crack images which is called NCD. Then, an adaptive bilateral filtering algorithm is developed, which can reduce the influence of noise. Finally, a new crack detection network with two new modules is designed. In the end, it is found that all the parts have promoting effects on crack detection under noise. Compared with other state‐of‐the‐art methods, this method performs better, especially in road crack detection under noise. When evaluating the well‐known crack500 test set, ODS F‐measure of 0.628 is achieved. Besides, this method is also evaluated in another five datasets. Significantly, ODS F‐measure of 0.545 is achieved, 4.0% higher than state‐of‐the‐art on GAPs384.

Previous studies have revealed the restorative effects of exposure to natural environments on psychological well-being and cognitive performance. Recent studies have reported the effects of exposure to nature sounds (e.g., the sounds of birds, rainfall, and waves) through a mobile application on reducing students' mental fatigue and improving their cognitive performance. However, it remains unknown whether exposure to nature sounds through a mobile application may influence students' learning performance. To address the gap, we conducted a study with 71 university students. During the four-week intervention, 36 students in the experimental group were exposed to nature sounds through a free mobile application for at least 30 consecutive minutes per day when working on academic-related tasks; 35 students in the control group did not have such exposure when working on similar tasks. The results show that students in the experimental group outperformed those in the control group in their engagement in deep learning, frequency of academic procrastination, and academic self-efficacy. The findings reveal the promising effects of exposure to nature sounds through a mobile application on improving students' learning performance. The implications of the findings are discussed.

Pixel-level road crack detection in UAV remote sensing images based on ARD-Unet

E-Learning and Massive Open Online Courses are old techniques, but since the Coronavirus, they have become more popular again. Students already suffer from a lack of concentration and motivation in traditional courses; thus, this lack affects online courses. Furthermore, another important Online Learning systems problem is the difference between learners in terms of Learning Styles, abilities, social characteristics as well as preferences, background, and other psychological and mental features. Generally, these features are not taken into account by scientists. Therefore, Deep Learning techniques and Datasets have been used to improve E-Learning systems and MOOCs in several aspects such as: predicting dropout, Learning Styles and performance of online learners, and even their attention after taking an online course. In this work, we have studied and analyzed many recent works in the area of using Deep Learning techniques to improve Online Learning systems and MOOCs. This analysis shows what researchers rely on to improve E-Learning and MOOCs and demonstrates that research does not use the definition of the appropriate Learning Style frequently. However, the most used ones are dropout and performance of learners. In another hand, learners' attention is still gap.

This paper comprehensively reviews the YOLOv8 algorithm and its application in road crack detection, focusing on the advantages of the algorithm in real-time target detection and high accuracy. With the continuous expansion of road infrastructure around the world, road cracks have become a serious safety hazard, which not only poses a threat to drivers but also brings a huge economic burden. Traditional road crack detection methods, such as manual inspection and sensor technology, often face the problems of high labor intensity, low efficiency and high cost. With the development of computer vision and deep learning technology, YOLOv8, as an efficient target detection system, provides new possibilities for the automation and accuracy improvement of road crack detection. This paper reviews different application versions of YOLOv8 in road crack detection, evaluates their performance, and points out the challenges of small target detection, real-time processing and multi-scale feature fusion. This paper proposes a variety of optimization strategies, including model optimization, data enhancement, training strategy and hardware acceleration. In addition, this paper looks forward to the future development trends of deep learning in the field of road crack detection, such as multimodal fusion, cross-domain detection and automated system integration, to promote the improvement of road safety and maintenance efficiency.

Facial expression recognition (FER) is a highly active field with applications in computer vision, human-computer interaction, security, and computer graphics animation. Recent advancements in deep learning and machine learning have increased interest in utilizing these techniques for accurate facial expression classification. This paper presents a comparative study that evaluates the performance of deep learning and machine learning as classifiers in FER systems, specifically after data fusion. Data fusion techniques combine and integrate multiple sources of information, aiming to enhance the overall classification accuracy by extracting two types of features using geometrical and appearance features trained using two types of convolutional neural networks. The feature outputs of these networks are fused to create a final feature vector for the classification process. The study evaluates the performance of deep learning on two benchmark datasets, the extended Cohn-Kanade (CK+) and Oulu-CASIA datasets, to assess the performance of deep learning. As a point of comparison, the traditional machine learning approach based on the support vector machine (SVM) is also evaluated on the same datasets. Performance metrics such as classification accuracy, precision, recall, and F1-score are utilized. The results obtained from the study highlight the strengths and limitations of both deep learning and machine learning techniques when employed as classifiers in FER systems. Notably, the experimental results demonstrate that the deep learning approach significantly outperforms the baseline methods, achieving an increase in recognition accuracy of 5.22% for the CK+ and 3.07% for the Oulu-CASIA dataset.

The study compares the diagnostic performance of deep learning (DL) with that of the former radiologist reading of the Kellgren–Lawrence (KL) grade and evaluates whether additional patient data can improve the diagnostic performance of DL. From March 2003 to February 2017, 3000 patients with 4366 knee AP radiographs were randomly selected. DL was trained using knee images and clinical information in two stages. In the first stage, DL was trained only with images and then in the second stage, it was trained with image data and clinical information. In the test set of image data, the areas under the receiver operating characteristic curve (AUC)s of the DL algorithm in diagnosing KL 0 to KL 4 were 0.91 (95% confidence interval (CI), 0.88–0.95), 0.80 (95% CI, 0.76–0.84), 0.69 (95% CI, 0.64–0.73), 0.86 (95% CI, 0.83–0.89), and 0.96 (95% CI, 0.94–0.98), respectively. In the test set with image data and additional patient information, the AUCs of the DL algorithm in diagnosing KL 0 to KL 4 were 0.97 (95% confidence interval (CI), 0.71–0.74), 0.85 (95% CI, 0.80–0.86), 0.75 (95% CI, 0.66–0.73), 0.86 (95% CI, 0.79–0.85), and 0.95 (95% CI, 0.91–0.97), respectively. The diagnostic performance of image data with additional patient information showed a statistically significantly higher AUC than image data alone in diagnosing KL 0, 1, and 2 (p-values were 0.008, 0.020, and 0.027, respectively).The diagnostic performance of DL was comparable to that of the former radiologist reading of the knee osteoarthritis KL grade. Additional patient information improved DL diagnosis in interpreting early knee osteoarthritis.

Road cracks are a common road traffic safety problem. Methods such as manual measurement of cracks do not facilitate large-scale inspections, which affects the normal operation of roads and the safety of pedestrians and vehicles. In this paper, a visual measurement technique based on convolutional neural networks and morphological operations is proposed for automated and efficient detection of road cracks. This method adopts the U-Net network as the infrastructure, changes the backbone feature extraction network to the VGG-16 network and introduces multiple indicators to establish a new loss function to alleviate the sample imbalance problem. Finally, the crack features are combined to perform morphological operations on the image to enrich the recovered detail features. After experiments on the road crack dataset, this method has better crack segmentation capability and reliability compared to other algorithms.

PurposeEducational data mining (EDM) discovers significant patterns from educational data and thus can help understand the relations between learners and their educational settings. However, most previous data mining techniques focus on prediction of learning performance of learners without integrating learning patterns identification techniques.Design/methodology/approachThis study proposes a new framework for identifying learning patterns and predicting learning performance. Two modules, the learning patterns identification module and the deep learning prediction models (DNN), are integrated into this framework to identify the difference of learning performance and predicting learning performance from profiles of students.FindingsExperimental results from survey data indicate that the proposed identifying learning patterns module could facilitate identifying valuable difference (change) patterns from student’s profiles. The proposed learning performance prediction module which adapts DNN also performs better than traditional machine techniques in prediction performance metrics.Originality/valueTo our best knowledge, the framework is the only educational system in the literature for identifying learning patterns and predicting learning performance.

The paper presents an exploration of the role of mobile technology for the realization of personalized learning and for improvement the students’ learning performance within an intelligent educational environment. The two of the created predictive models show the patterns and anomalies in students’ learning behavior and their learning performance. The utilized supervised machine learning algorithms: Random Forest, ID3, Naïve Bayes, Deep learning, k-NN are evaluated, and the results points out that the most suitable algorithms for solving these classification tasks are decision tree-based Random Forest and ID3. A multi-layer perceptron is used for predicting the students’ group learning performance at whole. KeywordsIntelligent educationLearning performanceMobile learningMachine learningNeural networks