scholarly journals Detection of wood surface defects based on improved YOLOv3 algorithm

BioResources ◽  
2021 ◽  
Vol 16 (4) ◽  
pp. 6766-6780
Author(s):  
Baogang Wang ◽  
Chunmei Yang ◽  
Yucheng Ding ◽  
Guangyi Qin
Keyword(s):  
Real Time ◽  
Wood Surface ◽  
Surface Defects ◽  
Detection Efficiency ◽  
Online Detection ◽  
Network Parameter ◽  
Baseline Model ◽  
Generalization Ability ◽  
The Mean ◽  
Residual Block

For the detection of wood surface defects, a convolutional neural network has a low detection efficiency and insufficient generalization ability, so it does not meet the requirements of online detection. Aiming to solve the above problems, the YOLOv3 baseline model, which has the advantage of multi-objective dynamic detection, was improved and applied to the online detection of wood surface defects. To solve the problem of the poor generalization ability of the network, GridMask was used to enhance the data and improve the robustness of the network. In order to solve the problem of the considerable amount of network parameter calculations and insufficient real-time performance, the residual block of the backbone network was changed to a Ghost block structure to achieve a lightweight model. Finally, the confidence loss function of the network was improved to reduce the influence of simple samples and negative samples on model convergence. The experimental results showed that, compared with the original network, the improved algorithm increased the mean average precision by 5.73% and the detection speed was increased to 28 frames per second (an increase of 11), which met the requirements for real-time industrial detection.

scholarly journals Surface Detection of Solid Wood Defects Based on SSD Improved with ResNet

Forests ◽  
2021 ◽  
Vol 12 (10) ◽  
pp. 1419
Author(s):  
Yutu Yang ◽  
Honghong Wang ◽  
Dong Jiang ◽  
Zhongkang Hu
Keyword(s):  
Wood Surface ◽  
Feature Detection ◽  
Surface Defects ◽  
Detection Efficiency ◽  
Solid Wood ◽  
Detection Accuracy ◽  
Single Shot ◽  
Wood Panel ◽  
Data Set ◽  
Wood Panels

Due to the lack of forest resources in China and the low detection efficiency of wood surface defects, the output of solid wood panels is not high. Therefore, this paper proposes a method for detecting surface defects of solid wood panels based on a Single Shot MultiBox Detector algorithm (SSD) to detect typical wood surface defects. The wood panel images are acquired by an independently designed image acquisition system. The SSD model included the first five layers of the VGG16 network, the SSD feature mapping layer, the feature detection layer, and the Non-Maximum Suppression (NMS) module. We used TensorFlow to train the network and further improved it on the basis of the SSD network structure. As the basic network part of the improved SSD model, the deep residual network (ResNet) replaced the VGG network part of the original SSD network to optimize the input features of the regression and classification tasks of the predicted bounding box. The solid wood panels selected in this paper are Chinese fir and pine. The defects include live knots, dead knots, decay, mildew, cracks, and pinholes. A total of more than 5000 samples were collected, and the data set was expanded to 100,000 through data enhancement methods. After using the improved SSD model, the average detection accuracy of the defects we obtained was 89.7%, and the average detection time was 90 ms. Both the detection accuracy and the detection speed were improved.

scholarly journals Electrical stimulator with mechanomyography-based real-time monitoring, muscle fatigue detection, and safety shut-off: a pilot study

2020 ◽  
Vol 65 (4) ◽  
pp. 461-468
Author(s):  
Jannatul Naeem ◽  
Nur Azah Hamzaid ◽  
Amelia Wong Azman ◽  
Manfred Bijak
Keyword(s):  
Muscle Fatigue ◽  
Real Time ◽  
Isometric Force ◽  
Mean Square ◽  
Clear Indication ◽  
Initial Value ◽  
Cord Injury ◽  
The Mean ◽  
Electrical Stimulator ◽  
Fatigue Detection

AbstractFunctional electrical stimulation (FES) has been used to produce force-related activities on the paralyzed muscle among spinal cord injury (SCI) individuals. Early muscle fatigue is an issue in all FES applications. If not properly monitored, overstimulation can occur, which can lead to muscle damage. A real-time mechanomyography (MMG)-based FES system was implemented on the quadriceps muscles of three individuals with SCI to generate an isometric force on both legs. Three threshold drop levels of MMG-root mean square (MMG-RMS) feature (thr50, thr60, and thr70; representing 50%, 60%, and 70% drop from initial MMG-RMS values, respectively) were used to terminate the stimulation session. The mean stimulation time increased when the MMG-RMS drop threshold increased (thr50: 22.7 s, thr60: 25.7 s, and thr70: 27.3 s), indicating longer sessions when lower performance drop was allowed. Moreover, at thr70, the torque dropped below 50% from the initial value in 14 trials, more than at thr50 and thr60. This is a clear indication of muscle fatigue detection using the MMG-RMS value. The stimulation time at thr70 was significantly longer (p = 0.013) than that at thr50. The results demonstrated that a real-time MMG-based FES monitoring system has the potential to prevent the onset of critical muscle fatigue in individuals with SCI in prolonged FES sessions.

scholarly journals Method for detecting surface defects of runner blades of large hydraulic turbines based on improved real-time lightweight network

2021 ◽  
Vol 1955 (1) ◽  
pp. 012090
Author(s):  
Cheng Liu ◽  
XinYi Su ◽  
Jialing Wu ◽  
Qun Zhou ◽  
Tao Li ◽  
...  
Keyword(s):  
Real Time ◽  
Surface Defects ◽  
Hydraulic Turbines

Lightweight detection algorithm for fine-grained surface defects of aerospace seal rings

2021 ◽  
pp. 1-18
Author(s):  
Hui Liu ◽  
Boxia He ◽  
Yong He ◽  
Xiaotian Tao
Keyword(s):  
Feature Extraction ◽  
Defect Detection ◽  
Surface Defect ◽  
Surface Defects ◽  
Detection Algorithm ◽  
Seal Ring ◽  
Fine Grained ◽  
Aerospace Applications ◽  
Surface Defect Detection ◽  
The Mean

The existing seal ring surface defect detection methods for aerospace applications have the problems of low detection efficiency, strong specificity, large fine-grained classification errors, and unstable detection results. Considering these problems, a fine-grained seal ring surface defect detection algorithm for aerospace applications is proposed. Based on analysis of the stacking process of standard convolution, heat maps of original pixels in the receptive field participating in the convolution operation are quantified and generated. According to the generated heat map, the feature extraction optimization method of convolution combinations with different dilation rates is proposed, and an efficient convolution feature extraction network containing three kinds of dilated convolutions is designed. Combined with the O-ring surface defect features, a multiscale defect detection network is designed. Before the head of multiscale classification and position regression, feature fusion tree modules are added to ensure the reuse and compression of the responsive features of different receptive fields on the same scale feature maps. Experimental results show that on the O-rings-3000 testing dataset, the mean condition accuracy of the proposed algorithm reaches 95.10% for 5 types of surface defects of aerospace O-rings. Compared with RefineDet, the mean condition accuracy of the proposed algorithm is only reduced by 1.79%, while the parameters and FLOPs are reduced by 35.29% and 64.90%, respectively. Moreover, the proposed algorithm has good adaptability to image blur and light changes caused by the cutting of imaging hardware, thus saving the cost.

scholarly journals Feasibility study of real-time online text-based CBT to support self-management for people with type 1 diabetes: the Diabetes On-line Therapy (DOT) Study

2021 ◽  
Vol 9 (1) ◽  
pp. e001934
Author(s):  
Anne M Doherty ◽  
Anne Herrmann-Werner ◽  
Arann Rowe ◽  
Jennie Brown ◽  
Scott Weich ◽  
...  
Keyword(s):  
Type 1 Diabetes ◽  
Glycemic Control ◽  
Real Time ◽  
Instant Messaging ◽  
Controlled Trial ◽  
Self Management ◽  
Diabetes Distress ◽  
The Mean ◽  
Distress Scale

IntroductionThis study examines the feasibility of conducting diabetes-focused cognitive–behavioral therapy (CBT) via a secure online real-time instant messaging system intervention to support self-management and improve glycemic control in people with type 1 diabetes.Research design and methodsWe used a pre–post uncontrolled intervention design over 12 months. We recruited adults with type 1 diabetes and suboptimal glycemic control (HbA1c ≥69 mmol/mol (DCCT 8.5%) for 12 months) across four hospitals in London. The intervention comprised 10 sessions of diabetes-focused CBT delivered by diabetes specialist nurses. The primary outcomes were number of eligible patients, rates of recruitment and follow-up, number of sessions completed and SD of the main outcome measure, change in HbA1c over 12 months. We measured the feasibility of collecting secondary outcomes, that is, depression measured using Patient Health Questionnaire-9 (PHQ-9), anxiety measured Generalised Anxiety Disorder (GAD) and the Diabetes Distress Scale (DDS).ResultsWe screened 3177 patients, of whom 638 were potentially eligible, from whom 71 (11.1%) were recruited. The mean age was 28.1 (13.1) years, and the mean HbA1c was 84.6 mmol/mol (17.8), DCCT 9.9%. Forty-six (65%) patients had at least 1 session and 29 (41%) completed all sessions. There was a significant reduction in HbA1c over 12 months (mean difference −6.2 (2.3) mmol/mol, DCCT 0.6%, p=0.038). The change scores in PHQ-9, GAD and DDS also improved.ConclusionsIt would be feasible to conduct a full-scale text-based synchronized real-time diabetes-focused CBT as an efficacy randomized controlled trial.

scholarly journals Utilization of laser interstitial thermotherapy guided by real-time thermal MRI as an alternative to separation surgery in the management of spinal metastasis

2015 ◽  
Vol 23 (4) ◽  
pp. 400-411 ◽  
Author(s):  
Claudio E. Tatsui ◽  
R. Jason Stafford ◽  
Jing Li ◽  
Jonathan N. Sellin ◽  
Behrang Amini ◽  
...  
Keyword(s):  
Quality Of Life ◽  
Spinal Cord ◽  
Real Time ◽  
Spinal Cord Compression ◽  
Spinal Metastasis ◽  
Cord Compression ◽  
The Mean ◽  
Interstitial Thermotherapy

OBJECT High-grade malignant spinal cord compression is commonly managed with a combination of surgery aimed at removing the epidural tumor, followed by spinal stereotactic radiosurgery (SSRS) aimed at local tumor control. The authors here introduce the use of spinal laser interstitial thermotherapy (SLITT) as an alternative to surgery prior to SSRS. METHODS Patients with a high degree of epidural malignant compression due to radioresistant tumors were selected for study. Visual analog scale (VAS) scores for pain and quality of life were obtained before and within 30 and 60 days after treatment. A laser probe was percutaneously placed in the epidural space. Real-time thermal MRI was used to monitor tissue damage in the region of interest. All patients received postoperative SSRS. The maximum thickness of the epidural tumor was measured, and the degree of epidural spinal cord compression (ESCC) was scored in pre- and postprocedure MRI. RESULTS In the 11 patients eligible for study, the mean VAS score for pain decreased from 6.18 in the preoperative period to 4.27 within 30 days and 2.8 within 60 days after the procedure. A similar VAS interrogating the percentage of quality of life demonstrated improvement from 60% preoperatively to 70% within both 30 and 60 days after treatment. Imaging follow-up 2 months after the procedure demonstrated a significant reduction in the mean thickness of the epidural tumor from 8.82 mm (95% CI 7.38–10.25) before treatment to 6.36 mm (95% CI 4.65–8.07) after SLITT and SSRS (p = 0.0001). The median preoperative ESCC Grade 2 was scored as 4, which was significantly higher than the score of 2 for Grade 1b (p = 0.04) on imaging follow-up 2 months after the procedure. CONCLUTIONS The authors present the first report on an innovative minimally invasive alternative to surgery in the management of spinal metastasis. In their early experience, SLITT has provided local control with low morbidity and improvement in both pain and the quality of life of patients.

scholarly journals Location accuracy of VLF World-Wide Lightning Location (WWLL) network: Post-algorithm upgrade

2005 ◽  
Vol 23 (2) ◽  
pp. 277-290 ◽  
Author(s):  
C. J. Rodger ◽  
J. B. Brundell ◽  
R. L. Dowden
Keyword(s):  
Real Time ◽  
World Wide ◽  
Detection Efficiency ◽  
Activity Levels ◽  
Operational Mode ◽  
Current Form ◽  
Location Accuracy ◽  
Central Processing ◽  
Lightning Discharges ◽  
Total Lightning

Abstract. An experimental VLF World-Wide Lightning Location (WWLL) network has been developed through collaborations with research institutions across the globe. The aim of the WWLL is to provide global real-time locations of lightning discharges, with >50% CG flash detection efficiency and mean location accuracy of <10km. While these goals are essentially arbitrary, they do define a point where the WWLL network development can be judged a success, providing a breakpoint for a more stable operational mode. The current network includes 18 stations which cover much of the globe. As part of the initial testing phase of the WWLL the network operated in a simple mode, sending the station trigger times into a central processing point rather than making use of the sferic Time of Group Arrival (TOGA). In this paper the location accuracy of the post-TOGA algorithm WWLL network (after 1 August 2003) is characterised, providing estimates of the globally varying location accuracy for this network configuration which range over 1.9-19km, with the global median being 2.9km, and the global mean 3.4km. The introduction of the TOGA algorithm has significantly improved the location accuracies. The detection efficiency of the WWLL is also considered. In the selected region the WWLL detected ~13% of the total lightning, suggesting a ~26% CG detection efficiency and a ~10% IC detection efficiency. Based on a comparison between all WWLL good lightning locations in February-April 2004, and the activity levels expected from satellite observations we estimate that the WWLL is currently detecting ~2% of the global total lightning, providing good locations for ~5% of global CG activity. The existing WWLL network is capable of providing real-time positions of global thunderstorm locations in its current form.

scholarly journals Efficient nested-PCR-based method development for detection and genotype identification of Acanthamoeba from a small volume of aquatic environmental sample

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Tsui-Kang Hsu ◽  
Jung-Sheng Chen ◽  
Hsin-Chi Tsai ◽  
Chi-Wei Tao ◽  
Yu-Yin Yang ◽  
...  
Keyword(s):  
Real Time ◽  
Real Time Pcr ◽  
Nested Pcr ◽  
In Silico ◽  
Environmental Samples ◽  
Method Development ◽  
Small Volume ◽  
Detection Efficiency ◽  
Genotyping Method ◽  
Pcr Method

AbstractAcanthamoeba spp. are opportunistic human pathogens that cause granulomatous amoebic encephalitis and keratitis, and their accurate detection and enumeration in environmental samples is a challenge. In addition, information regarding the genotyping of Acanthamoeba spp. using various PCR methods is equally critical. Therefore, considering the diverse niches of habitats, it is necessary to develop an even more efficient genotyping method for Acanthamoeba spp. detection. This study improved the sensitivity of detection to avoid underestimation of Acanthamoeba spp. occurrence in aquatic environmental samples, and to accurately define the pathogenic risk by developing an efficient PCR method. In this study, a new nested genotyping method was established and compared with various PCR-based methods using in silico, lab, and empirical tests. The in silico test showed that many PCR-based methods could not successfully align specific genotypes of Acanthamoeba, except for the newly designed nested PCR and real-time PCR method. Furthermore, 52 water samples from rivers, reservoirs, and a river basin in Taiwan were analysed by six different PCR methods and compared for genotyping and detection efficiency of Acanthamoeba. The newly developed nested-PCR-based method of genotyping was found to be significantly sensitive as it could effectively detect the occurrence of Acanthamoeba spp., which was underestimated by the JDP-PCR method. Additionally, the present results are consistent with previous studies indicating that the high prevalence of Acanthamoeba in the aquatic environment of Taiwan is attributed to the commonly found T4 genotype. Ultimately, we report the development of a small volume procedure, which is a combination of recent genotyping PCR and conventional real-time PCR for enumeration of aquatic Acanthamoeba and acquirement of biologically meaningful genotyping information. We anticipate that the newly developed detection method will contribute to the precise estimation, evaluation, and reduction of the contamination risk of pathogenic Acanthamoeba spp., which is regularly found in the water resources utilised for domestic purposes.

scholarly journals ORAL HIV-ASSOCIATED KAPOSI SARCOMA: A COMPARISON BETWEEN IMMUNOHISTOCHEMISTRY AND qPCR TECHNIQUES FOR DETECTION OF HHV8

2015 ◽  
Vol 21 (1) ◽  
pp. 29
Author(s):  
Priscila Lie Tobouti ◽  
Juliana Seo ◽  
Michella Bezerra Lima ◽  
Bruno Tavares Sedassari ◽  
Norberto Nobuo Sugaya ◽  
...  
Keyword(s):  
Real Time ◽  
Dna Extraction ◽  
Real Time Pcr ◽  
Kaposi Sarcoma ◽  
Positive Staining ◽  
Extraction Protocol ◽  
Simple Phenol ◽  
Two Samples ◽  
Hematoxylin And Eosin ◽  
The Mean

<p><strong>Objective: </strong>To compare the diagnostic accuracy of immunohistochemistry compared to real-time PCR (using a simple phenol-chloroform DNA extraction protocol) in the detection of HHV8 in small oral biopsies of Kaposi sarcoma. Also to validate the use of this DNA extraction protocol to extract HHV8 DNA.</p><p><strong>Material and methods:</strong> Seventeen cases of oral KS were examined. Data including gender, age, and anatomic location were obtained from patient´s records and histological sections stained with hematoxylin and eosin (H&amp;E) were reviewed. Immunohistochemistry was used to detect HHV8 in lesions of interest, as well as real-time PCR.</p><p><strong>Results: </strong>All the patients were HIV positive, the mean age was 35.5 years old, and the affected oral sites were palate (47%), gingiva (29.4%), tongue (11.8%), lip (5.9%), and oral floor (5.9%). Fifteen samples showed positive staining for HHV8 and only two samples were negative. The same results were observed using real-time PCR HHV8-DNA detection.</p><p><strong>Relevance: </strong>Our findings suggest that immunohistochemistry is faster and cheaper to perform than real-time PCR and was shown to have similar levels of sensitivity and accuracy for detection of HHV8 even in small biopsies. Additionally DNA extraction using a non-commercial kit, as done in this study can further reduce the costs to a pathology service.</p>

Sign in / Sign up

Export Citation Format

Share Document