Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

Machine vision is increasingly replacing manual steel surface inspection. The automatic inspection of steel surface defects makes it possible to ensure the quality of products in the steel industry with high accuracy. However, the optimization of inspection time presents a great challenge for the integration of machine vision in high-speed production lines. In this context, compressing the collected images before transmission is essential to save bandwidth and energy, and improve the latency of vision applications. The aim of this paper was to study the impact of quality degradation resulting from image compression on the classification performance of steel surface defects with a CNN. Image compression was applied to the Northeastern University (NEU) surface-defect database with various compression ratios. Three different models were trained and tested with these images to classify surface defects using three different approaches. The obtained results showed that trained and tested models on the same compression qualities maintained approximately the same classification performance for all used compression grades. In addition, the findings clearly indicated that the classification efficiency was affected when the training and test datasets were compressed using different parameters. This impact was more obvious when there was a large difference between these compression parameters, and for models that achieved very high accuracy. Finally, it was found that compression-based data augmentation significantly increased the classification precision to perfect scores (98–100%), and thus improved the generalization of models when tested on different compression qualities. The importance of this work lies in exploiting the obtained results to successfully integrate image compression into machine vision systems, and as appropriately as possible.

Rotation Estimation and Segmentation for Patterned Image Vision Inspection

Pattern images can be segmented in a template unit for efficient fabric vision inspection; however, segmentation criteria critically affect the segmentation and defect detection performance. To get the undistorted criteria for rotated images, rotation estimation of absolute angle needs to be proceeded. Given that conventional rotation estimations do not satisfy both rotation errors and computation times, patterned fabric defects are detected using manual visual methods. To solve these problems, this study proposes the application of segmentation reference point candidate (SRPC), generated based on a Euclidean distance map (EDM). SRPC is used to not only extract criteria points but also estimate rotation angle. The rotation angle is predicted using the orientation vector of SRPC instead of all pixels to reduce estimation times. SRPC-based image segmentation increases the robustness against the rotation angle and defects. The separation distance value for SRPC area distinction is calculated automatically. The performance of the proposed method is similar to state-of-the-art rotation estimation methods, with a suitable inspection time in actual operations for patterned fabric. The similarity between the segmented images is better than conventional methods. The proposed method extends the target of vision inspection on plane fabric to checked or striped pattern.

Comparison of GA, ACO algorithm, and PSO algorithm for path optimization on free-form surfaces using coordinate measuring machines

Coordinate measuring machines (CMMs) play an important role in modern manufacturing and inspection technologies. However, the inspection process of a CMM is recognized as time-consuming work. The low efficiency of coordinate measuring machines has given rise to new inspection strategies and methods, including path optimization. This study describes the optimization of an inspection path on free-form surfaces using three different algorithms: an ant colony optimization algorithm, a genetic algorithm, and a particle swarm optimization algorithm. The optimized sequence of sampling points is obtained in MATLAB R2020b software and tested on a Leitz Reference HP Bridge Type Coordinate Measuring Machine produced by HEXAGON. This study compares the performance of the three algorithms in theoretical and practical conditions. The results demonstrate that the use of the three algorithms can result in a collision-free path being found automatically and reduce the inspection time. However, owing to the different optimization methodologies, the optimized processes and optimized times of the three algorithms, as well as the optimized paths, are different. The results indicate that the ant colony algorithm has better performance for the path optimization of free-form surfaces.

Enhanced fluorescence signal using stray light shutter in a quantitative PCR chip

Quantitative polymerase chain reaction (qPCR) is the most important quantitative sensing technique for pathogens, especially for emerging pandemics such as coronavirus outbreak this year. The qPCR chip and device were investigated to meet the unmet needs of ultrafast inspection time, high accuracy, and small system volume. Therein, the fluorescence intensity was the most important signal in qPCR quantification of DNA amplifications, which is essential not only in the confirmative diagnosis of positive or negative infection, but also in the assessment of viral load for therapeutic and quarantine decision making. As the target DNAs got amplified, the interaction of fluorescence dye and double strand DNA will generate fluorescence signal proportional to amplified DNA in the intensity when excited by certain wavelength. A miniature spectro-detector was employed to receive the fluorescence scattering for digital output of the intensity in the qPCR chip in this study, and the optical simulation and actual experimental design and results according to the optical simulation results were performed to study the effect of the stray light shutter (SLS) in the improvement of the signal in fluorescence detection. The analysis results showed that the signal-to-noise ratio (SNR) of the fluorescence can be enhanced significantly for 5 times of the control using the SLS with a shape of extended component aperture, where the protruding structure was positioned away from the center. The experimental results showed that fluorescence intensity can be enhanced by 15.50% and 9.86% when adding the above shape of SLS in resin- and in glass-based chip, respectively. The results also demonstrated that the optical setup had good stability and repeatability in fluorescence detection, and variation was less than 1.00 %. Our results can provide important reference to the development of qPCR chip to obtain the high SNR fluorescence signal in DNA quantification process.

Gravimetric study and statistical data on the symbiotic effect of organic intermixtures on corrosion inhibition of P4 mold steel

The corrosion protection properties of the intermixed of l-leucine and vanillin (LLVL) on P4 mold steel within 1.5 M H2SO4 and HCl media was studied by weight loss analysis. Calculated data revealed the inhibition effect of the intermixed compound performed poorly at low H2SO4 concentrations due to inability of the combined inhibitor molecules to aggregate and effectively hinder the dispersion of the destructive anions to the steel surface. The inhibition efficiency decreased significantly with time to values below effective inhibitor performance. However, from mid to optimum concentration, inhibition efficiency of the compound was generally stable with time with values generally above 85%. In HCl media, the inhibition efficiency of the intermixture was generally above 80% at all concentrations with respect to the inspection time. The inhibitor compound exhibited greater stability in HCl compared to H2SO4 solution. Calculated values of standard deviation in H2SO4 were broadly greater than the outputs received in HCl due to the degree of variation between LLVL inhibition efficiency values. The margin of error at 95% confidence shows 65% of LLVL inhibition efficiency values obtained in H2SO4 solution have values above 70% inhibition efficiency with margin of error at +12.07% while in HCl solution, 100% of the LLVL inhibition efficiency data obtained is above 70% with margin of error of +1%. Analysis of variance showed the statistical relevance of inhibition efficiency in H2SO4 and HCl solution is significantly higher the corresponding relevance for inspection time with values of 70.45% and 71.18%.

Optimization of Inspection Plan for Transformer Verification Line Under Uninterrupted State

At present, transformer verification line of metering centre adopts fixed cycle inspection method manually. This method requires downtime for detection, which costs a lot of time and cost. Moreover, the inspection cycle is determined based on experience and lacks rigorous basis. To solve this problem, a hybrid delivery of inspection devices is proposed to realize non-stop detection and reduce the cost of inspection time. Considering impact of cost and false detection risk on inspection cycle, a multi-objective optimization model of inspection cycle based on inspection and false detection cost is proposed. Based on NSGA-II algorithm, perturbation population is introduced to enhance the global search ability, which aims to minimize the cost of inspection and false detection. Taking the verification line’s inspection plan of the metering centre as an example. It is solved by ENSGA-II algorithm, and feasibility of hybrid delivery mode is verified, which reduced downtime by 14.58%. A more reasonable inspection cycle is obtained, inspection cost is reduced by 29.57%, and false detection cost is reduced by 6.34%. It provides a reference for the formulation of inspection plan in the actual production process.

Comparison of Visual and Visual–Tactile Inspection of Aircraft Engine Blades

Background—In aircraft engine maintenance, the majority of parts, including engine blades, are inspected visually for any damage to ensure a safe operation. While this process is called visual inspection, there are other human senses encompassed in this process such as tactile perception. Thus, there is a need to better understand the effect of the tactile component on visual inspection performance and whether this effect is consistent for different defect types and expertise groups. Method—This study comprised three experiments, each designed to test different levels of visual and tactile abilities. In each experiment, six industry practitioners of three expertise groups inspected the same sample of N = 26 blades. A two-week interval was allowed between the experiments. Inspection performance was measured in terms of inspection accuracy, inspection time, and defect classification accuracy. Results—The results showed that unrestrained vision and the addition of tactile perception led to higher inspection accuracies of 76.9% and 84.0%, respectively, compared to screen-based inspection with 70.5% accuracy. An improvement was also noted in classification accuracy, as 39.1%, 67.5%, and 79.4% of defects were correctly classified in screen-based, full vision and visual–tactile inspection, respectively. The shortest inspection time was measured for screen-based inspection (18.134 s) followed by visual–tactile (22.140 s) and full vision (25.064 s). Dents benefited the most from the tactile sense, while the false positive rate remained unchanged across all experiments. Nicks and dents were the most difficult to detect and classify and were often confused by operators. Conclusions—Visual inspection in combination with tactile perception led to better performance in inspecting engine blades than visual inspection alone. This has implications for industrial training programmes for fault detection.

Test–Retest Reliability and Reliable Change Estimates for Four Mobile Cognitive Tests Administered Virtually in Community-Dwelling Adults

Objective: Remote mobile cognitive testing (MCT) is an expanding area of research, but psychometric data supporting these measures are limited. We provide preliminary data on test–retest reliability and reliable change estimates in four MCTs from SWAY Medical, Inc.Methods: Fifty-five adults from the U.S. Midwest completed the MCTs remotely on their personal mobile devices once per week for 3 consecutive weeks, while being supervised with a video-based virtual connection. The cognitive assessment measured simple reaction time (“Reaction Time”), go/no-go response inhibition (“Impulse Control”), timed visual processing (“Inspection Time”), and working memory (“Working Memory”). For each cognitive test except Working Memory, we analyzed both millisecond (ms) responses and an overall SWAY composite score.Results: The mean age of the sample was 26.69years (SD=9.89; range=18–58). Of the 55 adults, 38 (69.1%) were women and 49 (89.1%) used an iPhone. Friedman’s ANOVAs examining differences across testing sessions were nonsignificant (ps>0.31). Intraclass correlations for Weeks 1–3 were: Reaction Time (ms): 0.83, Reaction Time (SWAY): 0.83, Impulse Control (ms): 0.68, Impulse Control (SWAY): 0.80, Inspection Time (ms): 0.75, Inspection Time (SWAY): 0.75, and Working Memory (SWAY): 0.88. Intraclass correlations for Weeks 1–2 were: Reaction Time (ms): 0.75, Reaction Time (SWAY): 0.74, Impulse Control (ms): 0.60, Impulse Control (SWAY): 0.76, Inspection Time (ms): 0.79, Inspection Time (SWAY): 0.79, and Working Memory (SWAY): 0.83. Natural distributions of difference scores were calculated and reliable change estimates are presented for 70, 80, and 90% CIs.Conclusion: Test–retest reliability was adequate or better for the MCTs in this virtual remote testing study. Reliable change estimates allow for the determination of whether a particular level of improvement or decline in performance is within the range of probable measurement error. Additional reliability and validity data are needed in other age groups.

P109 The association between sleep microarchitecture and cognitive function in middle-aged and older men: A community-based study

Introduction Sleep microarchitecture metrics determined by quantitative power spectral analysis (PSA) of the electroencephalogram (EEG) have been proposed as potential biomarkers of cognitive function. However, there remain no data from community-based samples. This study examined cross-sectional associations between sleep microarchitecture metrics determined by PSA and cognitive function outcomes in community-dwelling men. Methods Men, Androgen, Inflammation, Lifestyle, Environment, and Stress (MAILES) study participants (n=477) underwent home-based polysomnography between 2010–2011. All-night EEG recordings were processed using PSA following exclusion of artefacts. MAILES participants also completed the inspection time task, Fuld object memory evaluation, and trail-making test A (TMT-A) and B (TMT-B). Multivariable linear regression models were used to determine the associations of sleep microarchitecture (relative spectral power) with cognitive function in the complete and age-stratified samples. Results Power spectral densities in theta-alpha ranges during NREM and REM sleep were associated with worse TMT-A performance, whereas higher delta density was associated with better TMT-A performance in the complete sample and men ≥65 years (all p<0.05). Similar associations were observed with TMT-B performance in men ≥65 years. Furthermore, in men <65 years, higher sigma density during NREM sleep was associated with faster inspection time (B= -3.14, 95% CI [-6.00, -0.27], p=0.032), whereas in men ≥65 years, higher theta density during NREM sleep was associated with faster inspection time (B = -3.33, 95% CI [-6.65, -0.02], p=0.049). Discussion PSA markers of sleep microarchitecture are independently associated with cognitive function. Longitudinal studies are needed to determine whether sleep microarchitecture metrics predict future cognitive dysfunction and decline.

P110 The association between sleep spindles and cognitive function in middle-aged and older men: A community-based study

Introduction The association between sleep spindles and cognitive function and the potential confounding influence of obstructive sleep apnea (OSA) remains uncertain. This study examined cross-sectional associations between sleep spindle metrics and cognitive function outcomes in community-dwelling men. Methods Men, Androgen, Inflammation, Lifestyle, Environment, and Stress (MAILES) study participants (n=477) underwent home-based polysomnography between 2010–2011 and completed the inspection time task, trail-making test A (TMT-A) and B (TMT-B), and Fuld object memory evaluation. Frontal spindle metrics derived from sleep electroencephalography included occurrence (total no. of sleep spindle events) and slow (11–13 Hz) and fast (13–16 Hz) spindle density (no./min) during N2 and N3 sleep. Results Men with OSA (any OSA and severe OSA) had significantly impaired sleep spindles (reduced occurrence and densities). In the complete study sample, higher spindle occurrence during N2 sleep was independently associated with faster inspection time (B= -0.44, 95% CI [-0.87, -0.02], p=0.041), whereas higher fast spindle density during N3 sleep was independently associated with worse TMT-B performance (B=20.7, 95% CI [0.55, 40.9], p=0.044). Furthermore, in men with severe OSA (apnea-hypopnea index ≥30/h), higher slow spindle density during N2 sleep was independently associated with worse TMT-A and TMT-B performance, whereas only higher spindle occurrence during N2 sleep was independently associated with worse TMT-A performance (all p<0.05). Discussion Specific spindle metrics during N2 and N3 sleep are independently associated with cognitive function in an unselected population of men and men with undiagnosed severe OSA. The utility of sleep spindles for predicting cognitive dysfunction and decline requires further investigation.