Automatic Bridge Crack Detection Using a Convolutional Neural Network

Concrete bridge crack detection is critical to guaranteeing transportation safety. The introduction of deep learning technology makes it possible to automatically and accurately detect cracks in bridges. We proposed an end-to-end crack detection model based on the convolutional neural network (CNN), taking the advantage of atrous convolution, Atrous Spatial Pyramid Pooling (ASPP) module and depthwise separable convolution. The atrous convolution obtains a larger receptive field without reducing the resolution. The ASPP module enables the network to extract multi-scale context information, while the depthwise separable convolution reduces computational complexity. The proposed model achieved a detection accuracy of 96.37% without pre-training. Experiments showed that, compared with traditional classification models, the proposed model has a better performance. Besides, the proposed model can be embedded in any convolutional network as an effective feature extraction structure.

Download Full-text

Voice activity detection in noisy conditions using tiny convolutional neural network

Informatics ◽

10.37661/1816-0301-2020-17-2-36-43 ◽

2020 ◽

Vol 17 (2) ◽

pp. 36-43

Author(s):

R. S. Vashkevich ◽

E. S. Azarov

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Model Performance ◽

Voice Activity Detection ◽

Detection Accuracy ◽

Activity Detection ◽

Detection Model ◽

Proposed Model ◽

Computational Resources ◽

Voice Activity

The paper investigates the problem of voice activity detection from a noisy sound signal. An extremely compact convolutional neural network is proposed. The model has only 385 trainable parameters. Proposed model doesn’t require a lot of computational resources that allows to use it as part of the “internet of things” concept for compact low power devices. At the same time the model provides state of the art results in voice activity detection in terms of detection accuracy. The properties of the model are achieved by using a special convolutional layer that considers the harmonic structure of vocal speech. This layer also eliminates redundancy of the model because it has invariance to changes of fundamental frequency. The model performance is evaluated in various noise conditions with different signal-to-noise ratios. The results show that the proposed model provides higher accuracy compared to voice activity detection model from the WebRTC framework by Google.

Download Full-text

Pavement Pothole Detection Based on Cascade and Fusion Convolutional Neural Network Using 2D Images under Complex Pavement Conditions

Recent Patents on Engineering ◽

10.2174/1872212114999200914113515 ◽

2020 ◽

Vol 14 ◽

Author(s):

Guoqiang Chen ◽

Bingxin Bai ◽

Huailong Yi

Keyword(s):

Neural Network ◽

Object Detection ◽

Convolutional Neural Network ◽

Feature Fusion ◽

Detection Accuracy ◽

Detection Model ◽

Automatic Driving ◽

Proposed Model ◽

Multi Level ◽

Pothole Detection

Background: Background: The development of deep learning technology has promoted the industrial intelligence, and automatic driving vehicles have become a hot research direction. As to the problem that pavement potholes threaten the safety of automatic driving vehicles, the pothole detection under complex environment conditions is studied. Objective: The goal of the work is to propose a new model of pavement pothole detection based on convolutional neural network. The main contribution is that the Multi-level Feature Fusion Block and the Detector Cascading Block are designed and a series of detectors are cascaded together to improve the detection accuracy of the proposed model. Methods: A pothole detection model is designed based on the original object detection model. In the study, the Transfer Connection Block in the Object Detection Module is removed and the Multi-level Feature Fusion Block is redesigned. At the same time, a Detector Cascading Block with multi-step detection is designed. Detectors are connected directly to the feature map and cascaded. In addition, the structure skips the transformation step. Results: The proposed method can be used to detect potholes efficiently. The real-time and accuracy of the model are improved after adjusting the network parameters and redesigning the model structure. The maximum detection accuracy of the proposed model is 75.24%. Conclusion: The Multi-level Feature Fusion Block designed enhances the fusion of high and low layer feature information and is conducive to extracting a large amount of target information. The Detector Cascade Block is a detector with cascade structure, which can realize more accurate prediction of the object. In a word, the model designed has greatly improved the detection accuracy and speed, which lays a solid foundation for pavement pothole detection under complex environmental conditions.

Download Full-text

A Deep Learning-Based Approach for the Detection of Early Signs of Gingivitis in Orthodontic Patients Using Faster Region-Based Convolutional Neural Networks

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph17228447 ◽

2020 ◽

Vol 17 (22) ◽

pp. 8447

Author(s):

Dima M. Alalharith ◽

Hajar M. Alharthi ◽

Wejdan M. Alghamdi ◽

Yasmine M. Alsenbel ◽

Nida Aslam ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Periodontal Disease ◽

Convolutional Neural Network ◽

Region Of Interest ◽

Training Dataset ◽

Detection Accuracy ◽

Detection Model ◽

Art Object ◽

Orthodontic Patients

Computer-based technologies play a central role in the dentistry field, as they present many methods for diagnosing and detecting various diseases, such as periodontitis. The current study aimed to develop and evaluate the state-of-the-art object detection and recognition techniques and deep learning algorithms for the automatic detection of periodontal disease in orthodontic patients using intraoral images. In this study, a total of 134 intraoral images were divided into a training dataset (n = 107 [80%]) and a test dataset (n = 27 [20%]). Two Faster Region-based Convolutional Neural Network (R-CNN) models using ResNet-50 Convolutional Neural Network (CNN) were developed. The first model detects the teeth to locate the region of interest (ROI), while the second model detects gingival inflammation. The detection accuracy, precision, recall, and mean average precision (mAP) were calculated to verify the significance of the proposed model. The teeth detection model achieved an accuracy, precision, recall, and mAP of 100 %, 100%, 51.85%, and 100%, respectively. The inflammation detection model achieved an accuracy, precision, recall, and mAP of 77.12%, 88.02%, 41.75%, and 68.19%, respectively. This study proved the viability of deep learning models for the detection and diagnosis of gingivitis in intraoral images. Hence, this highlights its potential usability in the field of dentistry and aiding in reducing the severity of periodontal disease globally through preemptive non-invasive diagnosis.

Download Full-text

Pavement Crack Detection and Segmentation Method Based on Improved Deep Learning Fusion Model

Mathematical Problems in Engineering ◽

10.1155/2020/8515213 ◽

2020 ◽

Vol 2020 ◽

pp. 1-22

Author(s):

Xiaoran Feng ◽

Liyang Xiao ◽

Wei Li ◽

Lili Pei ◽

Zhaoyun Sun ◽

...

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Geometric Parameter ◽

Crack Detection ◽

Crack Identification ◽

Single Shot ◽

Road Maintenance ◽

Fusion Model ◽

Detection Model ◽

Crack Classification

Pavement damage is the main factor affecting road performance. Pavement cracking, a common type of road damage, is a key challenge in road maintenance. In order to achieve an accurate crack classification, segmentation, and geometric parameter calculation, this paper proposes a method based on a deep convolutional neural network fusion model for pavement crack identification, which combines the advantages of the multitarget single-shot multibox detector (SSD) convolutional neural network model and the U-Net model. First, the crack classification and detection model is applied to classify the cracks and obtain the detection confidence. Next, the crack segmentation network is applied to accurately segment the pavement cracks. By improving the feature extraction structure and optimizing the hyperparameters of the model, pavement crack classification and segmentation accuracy were improved. Finally, the length and width (for linear cracks) and the area (for alligator cracks) are calculated according to the segmentation results. Test results show that the recognition accuracy of the pavement crack identification method for transverse, longitudinal, and alligator cracks is 86.8%, 87.6%, and 85.5%, respectively. It is demonstrated that the proposed method can provide the category information for pavement cracks as well as the accurate positioning and geometric parameter information, which can be used directly for evaluating the pavement condition.

Download Full-text

Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection

Sensors ◽

10.3390/s20164403 ◽

2020 ◽

Vol 20 (16) ◽

pp. 4403

Author(s):

Umme Hafsa Billah ◽

Hung Manh La ◽

Alireza Tavakkoli

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Network Architecture ◽

Crack Detection ◽

Inspection System ◽

Detection Accuracy ◽

Feature Maps ◽

Neural Network Architecture ◽

Concrete Crack ◽

Discriminative Feature

An autonomous concrete crack inspection system is necessary for preventing hazardous incidents arising from deteriorated concrete surfaces. In this paper, we present a concrete crack detection framework to aid the process of automated inspection. The proposed approach employs a deep convolutional neural network architecture for crack segmentation, while addressing the effect of gradient vanishing problem. A feature silencing module is incorporated in the proposed framework, capable of eliminating non-discriminative feature maps from the network to improve performance. Experimental results support the benefit of incorporating feature silencing within a convolutional neural network architecture for improving the network’s robustness, sensitivity, and specificity. An added benefit of the proposed architecture is its ability to accommodate for the trade-off between specificity (positive class detection accuracy) and sensitivity (negative class detection accuracy) with respect to the target application. Furthermore, the proposed framework achieves a high precision rate and processing time than the state-of-the-art crack detection architectures.

Download Full-text

Convolutional neural network for automated mass segmentation in mammography

BMC Bioinformatics ◽

10.1186/s12859-020-3521-y ◽

2020 ◽

Vol 21 (S1) ◽

Author(s):

Dina Abdelhafiz ◽

Jinbo Bi ◽

Reda Ammar ◽

Clifford Yang ◽

Sheida Nabavi

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Data Augmentation ◽

Contextual Information ◽

Automatic Segmentation ◽

Ground Truth ◽

Test Accuracy ◽

Convolutional Network ◽

Proposed Model ◽

The Mean

Abstract Background Automatic segmentation and localization of lesions in mammogram (MG) images are challenging even with employing advanced methods such as deep learning (DL) methods. We developed a new model based on the architecture of the semantic segmentation U-Net model to precisely segment mass lesions in MG images. The proposed end-to-end convolutional neural network (CNN) based model extracts contextual information by combining low-level and high-level features. We trained the proposed model using huge publicly available databases, (CBIS-DDSM, BCDR-01, and INbreast), and a private database from the University of Connecticut Health Center (UCHC). Results We compared the performance of the proposed model with those of the state-of-the-art DL models including the fully convolutional network (FCN), SegNet, Dilated-Net, original U-Net, and Faster R-CNN models and the conventional region growing (RG) method. The proposed Vanilla U-Net model outperforms the Faster R-CNN model significantly in terms of the runtime and the Intersection over Union metric (IOU). Training with digitized film-based and fully digitized MG images, the proposed Vanilla U-Net model achieves a mean test accuracy of 92.6%. The proposed model achieves a mean Dice coefficient index (DI) of 0.951 and a mean IOU of 0.909 that show how close the output segments are to the corresponding lesions in the ground truth maps. Data augmentation has been very effective in our experiments resulting in an increase in the mean DI and the mean IOU from 0.922 to 0.951 and 0.856 to 0.909, respectively. Conclusions The proposed Vanilla U-Net based model can be used for precise segmentation of masses in MG images. This is because the segmentation process incorporates more multi-scale spatial context, and captures more local and global context to predict a precise pixel-wise segmentation map of an input full MG image. These detected maps can help radiologists in differentiating benign and malignant lesions depend on the lesion shapes. We show that using transfer learning, introducing augmentation, and modifying the architecture of the original model results in better performance in terms of the mean accuracy, the mean DI, and the mean IOU in detecting mass lesion compared to the other DL and the conventional models.

Download Full-text

Automated stomata detection in oil palm with convolutional neural network

Scientific Reports ◽

10.1038/s41598-021-94705-4 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Qi Bin Kwong ◽

Yick Ching Wong ◽

Phei Ling Lee ◽

Muhammad Syafiq Sahaini ◽

Yee Thung Kon ◽

...

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Oil Palm ◽

Stomatal Density ◽

Density Measurement ◽

Automated System ◽

Detection Accuracy ◽

Detection Model ◽

Drought Tolerant ◽

Breeding Selection

AbstractStomatal density is an important trait for breeding selection of drought tolerant oil palms; however, its measurement is extremely tedious. To accelerate this process, we developed an automated system. Leaf samples from 128 palms ranging from nursery (1 years old), juvenile (2–3 years old) and mature (> 10 years old) were collected to build an oil palm specific stomata detection model. Micrographs were split into tiles, then used to train a stomata object detection convolutional neural network model through transfer learning. The detection model was then tested on leaf samples acquired from three independent oil palm populations of young seedlings (A), juveniles (B) and productive adults (C). The detection accuracy, measured in precision and recall, was 98.00% and 99.50% for set A, 99.70% and 97.65% for set B, and 99.55% and 99.62% for set C, respectively. The detection model was cross-applied to another set of adult palms using stomata images taken with a different microscope and under different conditions (D), resulting in precision and recall accuracy of 99.72% and 96.88%, respectively. This indicates that the model built generalized well, in addition has high transferability. With the completion of this detection model, stomatal density measurement can be accelerated. This in turn will accelerate the breeding selection for drought tolerance.

Download Full-text

Image recognition algorithms based on deep learning

Journal of Physics Conference Series ◽

10.1088/1742-6596/2137/1/012056 ◽

2021 ◽

Vol 2137 (1) ◽

pp. 012056

Author(s):

Hongli Ma ◽

Fang Xie ◽

Tao Chen ◽

Lei Liang ◽

Jie Lu

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Convolutional Neural Network ◽

Convolutional Neural Networks ◽

Development Process ◽

Research Direction ◽

Learning Technology ◽

Convolutional Network ◽

Important Research Direction

Abstract Convolutional neural network is a very important research direction in deep learning technology. According to the current development of convolutional network, in this paper, convolutional neural networks are induced. Firstly, this paper induces the development process of convolutional neural network; then it introduces the structure of convolutional neural network and some typical convolutional neural networks. Finally, several examples of the application of deep learning is introduced.

Download Full-text

Detection Of Vacant Parking Spaces Through The Use Of Convolutional Neural Network

The International FLAIRS Conference Proceedings ◽

10.32473/flairs.v34i1.128470 ◽

2021 ◽

Vol 34 (1) ◽

Author(s):

Christopher Singh ◽

Christoforos Christoforou

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Transfer Learning ◽

Automotive Industry ◽

Time Of Day ◽

Detection Accuracy ◽

Driving Time ◽

Parking Spot ◽

Proposed Model ◽

Time Required

This paper focuses on the application of computer vision and convolutional neural network techniques in the automotive industry to reduce the amount of time required to locate a vacant parking spot and to reduce driving time. The main motivation for a vacant parking spot detector is such that today’s drivers are facing major difficulties in finding available spots in largely populated cities. This often time leads to increased congestion and frustration for the driver because they are forced to continue their search for a parking spot. Our approach is able to solve this issue and provide the driver with useful information through the use of transfer learning methodologies. The main contribution of this paper is to examine and improve on previously implemented transfer learning methods in order to better increase the detection accuracy. This paper differs from previous attempts such that it considers all environmental factors such as weather and time of day. Other models are not able to handle these conditions with a high accuracy and subsequently falter. When compared to previous attempts, our implementation focuses solely on the reliance of transfer learning. The results indicate that our model is capable of identifying vacant parking spaces under all conditions with competitive accuracies. The proposed model is able to surpass the accuracy of the latest attempt at solving this issue.

Download Full-text

Batik Classification using Deep Convolutional Network Transfer Learning

Jurnal Ilmu Komputer dan Informasi ◽

10.21609/jiki.v11i2.507 ◽

2018 ◽

Vol 11 (2) ◽

pp. 59 ◽

Cited By ~ 5

Author(s):

Yohanes Gultom ◽

Aniati Murni Arymurthy ◽

Rian Josua Masikome

Keyword(s):

Neural Network ◽

Deep Learning ◽

Cultural Heritage ◽

Convolutional Neural Network ◽

Transfer Learning ◽

Data Representation ◽

Convolutional Network ◽

Proposed Model ◽

Feature Extractor ◽

Research Task

Batik fabric is one of the most profound cultural heritage in Indonesia. Hence, continuous research on understanding it is necessary to preserve it. Despite of being one of the most common research task, Batik’s pattern automatic classification still requires some improvement especially in regards to invariance dilemma. Convolutional neural network (ConvNet) is one of deep learning architecture which able to learn data representation by combining local receptive inputs, weight sharing and convolutions in order to solve invariance dilemma in image classification. Using dataset of 2,092 Batik patches (5 classes), the experiments show that the proposed model, which used deep ConvNet VGG16 as feature extractor (transfer learning), achieves slightly better average of 89 ± 7% accuracy than SIFT and SURF-based that achieve 88 ± 10% and 88 ± 8% respectively. Despite of that, SIFT reaches around 5% better accuracy in rotated and scaled dataset.

Download Full-text