Quality Classification Method Research Articles

Comprehensive Clinical Usability-Oriented Contour Quality Evaluation for Deep Learning Auto-segmentation: Combining Multiple Quantitative Metrics Through Machine Learning

PurposeThe current commonly-used metrics for evaluating the quality of auto-segmented contours have limitations and do not always reflect the clinical usefulness of the contours. This work aims to develop a novel contour quality classification (CQC) method by combining multiple quantitative metrics for clinical usability-oriented contour quality evaluation for deep learning-based auto-segmentation (DLAS). MethodsThe CQC was designed to categorize contours on slices as acceptable, minor edit, or major edit based on the expected editing effort/time with supervised ensemble tree classification models using seven quantitative metrics. Organ-specific models were trained for five abdominal organs (pancreas, duodenum, stomach, small and large-bowels) using 50 MRI datasets. Twenty additional MRI and nine CT datasets were employed for testing. Inter-observer variation (IOV) was assessed among six observers and consensus labels were established through majority vote for evaluation. The CQC was also compared with a threshold-based baseline approach. ResultsFor the five organs, the average AUC was 0.982±0.01 and 0.979±0.01, the mean-accuracy was 95.8±1.7% and 94.3±2.1%, and the mean risk-rate was 0.8±0.4% and 0.7±0.5% for MRI and CT testing dataset, respectively. The CQC results closely matched the IOV results (mean-accuracy of 94.2±0.8% and 94.8±1.7%) and were significantly higher than those obtained using the threshold-based method (mean-accuracy of 80.0±4.7%, 83.8±5.2%, and 77.3±6.6% using one, two, and three metrics). ConclusionThe CQC models demonstrated high performance in classifying the quality of contour slices. This method can address the limitations of existing metrics and offers an intuitive and comprehensive solution for clinically oriented evaluation and comparison of DLAS systems.

An appearance quality classification method for Auricularia auricula based on deep learning

The intelligent appearance quality classification method for Auricularia auricula is of great significance to promote this industry. This paper proposes an appearance quality classification method for Auricularia auricula based on the improved Faster Region-based Convolutional Neural Networks (improved Faster RCNN) framework. The original Faster RCNN is improved by establishing a multiscale feature fusion detection model to improve the accuracy and real-time performance of the model. The multiscale feature fusion detection model makes full use of shallow feature information to complete target detection. It fuses shallow features with rich detailed information with deep features rich in strong semantic information. Since the fusion algorithm directly uses the existing information of the feature extraction network, there is no additional calculation. The fused features contain more original detailed feature information. Therefore, the improved Faster RCNN can improve the final detection rate without sacrificing speed. By comparing with the original Faster RCNN model, the mean average precision (mAP) of the improved Faster RCNN is increased by 2.13%. The average precision (AP) of the first-level Auricularia auricula is almost unchanged at a high level. The AP of the second-level Auricularia auricula is increased by nearly 5%. And the third-level Auricularia auricula AP is increased by 1%. The improved Faster RCNN improves the frames per second from 6.81 of the original Faster RCNN to 13.5. Meanwhile, the influence of complex environment and image resolution on the Auricularia auricula detection is explored.

MFCC-based Audio Quality Classification Method for Forensics

MFCC-based Audio Quality Classification Method for Forensics

Implementation of the Ordinal Logistic Regression Method for Air Quality Classification Based on the Air Pollution Standard Index

Air pollution is a serious problem in many cities around the world, caused by human and natural activities. Jakarta, as an Indonesian economic and transportation hub, faces serious air pollution challenges. This study uses the Ordinal Logistics Regression Method to develop an optimal classification model for identifying air quality based on the air pollution index. The aim is to contribute to dealing with air contamination and improve understanding of the use of such methods in the classification of air quality. Data used from 2012 to 2021 covered parameters such as PM10, SO2, CO, O3, and NO2. Oversampling is done using SMOTE to address imbalances in the datasets used. The data used is divided into two parts: training data and validation test data, with an 80:20 ratio. The model training and testing process is carried out with a variety of scenarios, including parameter significance tests using probability tests and Wald tests, cross-validation with fold numbers 5, 10, and 15, as well as evaluation using a confusion matrix. Models are used for Mord libraries such as Ordinalridge, LogisticAT, logisticIT, and LogisticSE. There were a total of 72 model testing experiments to find the best model of the six data model outputs. The test results showed that the optimal model was obtained with a data deletion scenario of null values, data oversampling using the LogisticIT model, and K-fold = 5, with a training accuracy of 0.8628 and validation data test accuracy of 0.8599, as well as each precision value of 0.86, recall value of 0.86, and F1-score of 0.86. The model's performance was satisfactory in handling different data variations, according to the evaluation. These results show that the model is able to generalize data well and is reliable in predicting air quality accurately.

Application of K-PSO Clustering Algorithm and Game Theory in Rock Mass Quality Evaluation of Maji Hydropower Station

In this study, the K-means algorithm based on particle swarm optimization (K-PSO) and game theory are introduced to establish the quality evaluation model of a rock mass. Five evaluation factors were considered, i.e., uniaxial saturated compressive strength of rock, discontinuity spacing, acoustic velocity, rock quality designation (RQD), and integrity coefficient. The rock mass of an elevation adit at the abutment of Maji hydropower station was taken as a case study. The subjective weight of the evaluation factor was determined by the weighted least squares method, and the objective weight of the evaluation factor was determined by the entropy method. The combined weights of each influencing factor were determined by game theory to be 0.142, 0.179, 0.035, 0.116, and 0.108. The rock mass quality evaluation in the study area was analyzed by K-PSO algorithm. The results indicate that the K-PSO clustering results are almost the same as the evaluation results of the traditional basic quality (BQ) classification method and the widely used extension evaluation method and are consistent with the preliminary judgment of the expert field. The results are consistent with the field observation law. It is considered that the K-PSO clustering theory can reflect the engineering geological characteristics of the rock mass of the hydropower project in the rock mass quality evaluation.

Design and Experimental Research of Intelligent Inspection and Classification System for Yuba Skin Quality

At present, the surface quality of Yuba skin is determined by sensory methods. In order to realize the intelligent classification detection of Yuba skin quality, this study designed a system that automatically determines the quality of Yuba skin surfaces based on image processing and support vector machine (SVM) approaches. Specifically, the system uses image preprocessing to extract the grayscale eigenvalues, gray level co-occurrence matrix (GLCM) eigenvalues, and gray level run length matrix (GLRLM) eigenvalues of the sample image and uses them as input values for a quality grading system. Through model evaluation of three classification models, the SVM classification model was selected according to the evaluation results, and different kernel functions were used in the model for sample training. Based on Matlab, the quality grading software of Yuba skin was developed and designed. Intelligent detection and grading were realized through the radial basis kernel function support vector machine (RBF-SVM) grading model. The best penalty factor (c = 3.50) and kernel parameter value (g = 0.98) were obtained through cross-validation. The accuracy of the model was 95.31% and 94.16% for the training and test sets, respectively. The grading accuracy of the RBF-SVM grading system was 93.56%, and the error was less than 5% compared with the traditional sensory method of grading; thus, the quality classification method based on the SVM classification system for Yuba skin is feasible and can be used for quality detection.

Conv-Tire: Tire Condition Assessment using Convolutional Neural Networks

Purpose: In this study, the authors designed an algorithm based on convolutional neural networks that can automatically assess tire quality.Design/methodology/approach: The proposed algorithm is built through several stages as follows. In the first stage, the tire images, which are the input of the designed algorithm, are acquired. Further, the acquired images are divided into two sets, namely training and testing sets. The training set contains tire images used in the training phase of several convolutional neural networks (CNN) architectures such as ResNet-50, MobileNetV2, Inception V3, and DenseNet-121. The training phase is carried out in a number of epochs, and at each epoch, the cross entropy loss function will be calculated which expresses the performance of the CNN architecture in classifying tire images. For this reason, the training stage requires a label or reference that shows the feasibility of the tires displayed in each image.Findings/result: In the testing phase, trained CNN architectures are used to classify tire images from the test set. Classification performance in the test set is also expressed in terms of cross-entropy loss function value. In addition, the accuracy value has also been calculated which shows the percentage of the number of tire images that are successfully classified correctly to the total number of tire images in the test set, namely the DenseNet-121 model has the best accuracy of 92.62%.Originality/value/state of the art: Given the high accuracy achieved by our algorithm, this work can be used as a reference by other researchers, specifically to benchmark their tire quality classification methods developed in the future.

Appearance quality classification method of Huangguan pear under complex background based on instance segmentation and semantic segmentation.

The 'Huangguan' pear disease spot detection and grading is the key to fruit processing automation. Due to the variety of individual shapes and disease spot types of 'Huangguan' pear. The traditional computer vision technology and pattern recognition methods have some limitations in the detection of 'Huangguan' pear diseases. In recent years, with the development of deep learning technology and convolutional neural network provides a new solution for the fast and accurate detection of 'Huangguan' pear diseases. To achieve automatic grading of 'Huangguan' pear appearance quality in a complex context, this study proposes an integrated framework combining instance segmentation, semantic segmentation and grading models. In the first stage, Mask R-CNN and Mask R-CNN with the introduction of the preprocessing module are used to segment 'Huangguan' pears from complex backgrounds. In the second stage, DeepLabV3+, UNet and PSPNet are used to segment the 'Huangguan' pear spots to get the spots, and the ratio of the spot pixel area to the 'Huangguan' pear pixel area is calculated and classified into three grades. In the third stage, the grades of 'Huangguan' pear are obtained using ResNet50, VGG16 and MobileNetV3. The experimental results show that the model proposed in this paper can segment the 'Huangguan' pear and disease spots in complex background in steps, and complete the grading of 'Huangguan' pear fruit disease severity. According to the experimental results. The Mask R-CNN that introduced the CLAHE preprocessing module in the first-stage instance segmentation model is the most accurate. The resulting pixel accuracy (PA) is 97.38% and the Dice coefficient is 68.08%. DeepLabV3+ is the most accurate in the second-stage semantic segmentation model. The pixel accuracy is 94.03% and the Dice coefficient is 67.25%. ResNet50 is the most accurate among the third-stage classification models. The average precision (AP) was 97.41% and the F1 (harmonic average assessment) was 95.43%.In short, it not only provides a new framework for the detection and identification of 'Huangguan' pear fruit diseases in complex backgrounds, but also lays a theoretical foundation for the assessment and grading of 'Huangguan' pear diseases.

Water Quality Classification for Inland Lakes and Ponds with Few Color Image Samples Based on Triple-GAN and CSNN

Water color is an important representation reflecting the characteristics of its quality in inland lakes or ponds; however, sufficient water color image samples are often difficult to obtain due to the limitation of fishery production. For few color image samples, the existing data enhancement methods based on the depth generation model have the problems of low quality of generated data, difficulty of network training, and so on; moreover, for image classification, traditional methods based on convolutional neural network (CNN) cannot effectively extract the potential manifold structure features in the image and the full connection layer in CNN cannot simulate biological neurons well, resulting in high time cost and low efficiency. In this paper, a water quality classification method has been proposed to solve the above problems, the improved semisupervised triple-generation adversarial network (triple-GAN) algorithm is used to enhance the few water color image samples, and the feature data can then be extracted from enhanced data by manifold learning method t-distributed stochastic neighborhood embedding (t-SNE). Moreover, convolutional spiking neural network (CSNN), in which spiking neural network (SNN) has replaced the original full connection layer of CNN, is used for final water quality classification. The main contribution of this paper is to build a new algorithm framework, introduce triple-GAN and CSNN into the field of classification of few water color image samples for the first time, and make an exploration of integrating artificial intelligence (AI) and water quality analysis problems. By comparing with traditional methods, the proposed method is proved to have the advantages of less time-consuming, low operation cost, and high classification accuracy.

Apple appearance quality classification method based on double branch feature fusion network

This paper carried out a classification study on the quality of different types of apples. In order to achieve a high-precision classification of high-quality apples in various dimensions, shorten the classification time, improve efficiency, and make it feasible for practical applications. This paper proposes an automatic recognition and classification model OB-Net (for apple appearance quality) based on a dual-branch structure. The model consists of two branches, an O branch and a B branch. The O branch extracts the shape and contour features of the apple surface defect by decomposing the feature map into high and low frequencies. The B branch extracts the shape size and texture characteristics of the apple surface defect by fusing channel attention and spatial attention mechanisms; this is done to further increase the feature distance between different types of defects. Experimental results show that the classification accuracy rates of rot, insect bites, russeting, scratches, and intact apples are 95.65%, 98.17%, 94.62%, 92.02% and 97.67%, respectively; the overall classification and recognition accuracy rate reaches 95.64%. Finally, the three aspects of the feature map, heat map and category probability statistics map extracted from the OB-Net network model prove the validity of the network model in the classification of apple appearance quality.

An efficient IoT based prediction system for classification of water using novel adaptive incremental learning framework

Creating an adaptive, accurate, and reliable model is a universal problem. Machine learning models give poor accuracy on unseen data, and therefore, the testing accuracy of the trained model is affected. This study presents a novel adaptive incremental learning framework for IoT based smart water quality classification system to predict the suitability of water for different applications. Initially, water quality data is collected using IoT sensors. After that, data cleaning is performed by removing missing values and outliers. Next, features associated with the sensed data are obtained, and unwanted features are removed. Then, the G-SMOTE technique is proposed, which hybridizes the SMOTE and the genetic algorithm to address the imbalanced data set problem. After that, the multi-class classification is performed using a modified deep learning neural network classifier which uses hyperparameter tunning technique to obtain better accuracy with minimum validation loss. Finally, the study presents a novel framework for adaptive incremental learning on unseen data. Experimental result shows that our method presents a new state-of-the-art multi-class water quality classification method with an accuracy of 99.34% and validation loss of 0.0415.

Classification Method of Rock Structure and Rock Mass Quality of Surface Granite: Geological Disposal of High-Level Radioactive Waste in China

The engineering quality of rock mass is a key factor to evaluate the long-term stability and safety of high-level radioactive waste (HLW) geological disposal engineering and is also the important basis for disposal site selection. Traditional rock mass quality classification methods, such as RMR and [Formula: see text], can meet underground engineering but still should be studied further for the site evaluation in the HLW disposal engineering. In this study, rock mass structure rating (RMSG) was proposed based on the quantitative control index of rock mass structure which was from the rock mass quality classification methods. Based on the statistical results of rock mass structure, the relationship between the number of RMSG and the modified RMR ([Formula: see text]) and [Formula: see text] ([Formula: see text]) was established, China, as a case study. Results from this study show that RMSG is linearly related to [Formula: see text] and negatively exponential to [Formula: see text]. The research results can solve the evaluation of rock mass quality for HLW geological disposal engineering, and the addition of more engineering examples over time will enable further verification.

Fuzzy system modelling to assess water quality for irrigation purposes

This study attempts to find a fuzzy logic system for assessing the quality of water in water treatment plants (WTPs) providing water for irrigation purposes in the Basrah Governorate (South of Iraq). Each month, samples are taken in each of six major WTPs to measure electrical conductivity ( EC), and the content of sodium, magnesium and calcium. The calculated value which is the sodium adsorption ratio ( SAR) is plotted with EC on the Richard diagram. SAR and EC values are combined together in a fuzzy inference system (FIS) to find out a quality number called the fuzzy irrigation water quality index number ( FIWQI) which ranges from zero to one. The higher the value of the index, the better water quality. The Richard diagram, which helps to classify irrigation water, is used to adjust FIS components. Results show that the FIWQI for all WTPs changes depending on location and season. It ranges between 0.114–0.170, 0.120–0.190, 0.114–0.170, 0.114–0.202, 0.118–0.500 and 0.46–0.500 for Al-Bradhaia 1, Al-Jubaila 1, Shatt Al-Arab, Garmmah 1, Al-Rebat, and Old Shauaibah WTPs, respectively. The results indicate that WTPs effluent drawn from the Shatt Al-Arab River has poor water quality for irrigation purposes, except for an Old Shauaibah which receives water from another source called a sweet water canal. FIS results are compared with values obtained from the Richard diagram and 96% degree of compatibility between the two methods is attained. This indicates that FIS is an acceptable method for water quality classification.

Research on Test and Logging Data Quality Classification for Gas–Water Identification

Tight sandstone oil and gas reservoirs are widely distributed, rich in resources, with a bright prospect for exploration and development in China. Due to multiple evolutions of the structure and sedimentary system, the gas–water distribution laws are complicated in tight sandstone gas reservoirs in the northern Ordos area. It is difficult to identify gas and water layers in the study area. In addition, in the development and production, various factors, such as the failure of the instrument, the difference in construction parameters (injected sand volume, flowback rate), poor test results, and multi-layer joint testing lead to unreliable gas test results. Then, the inaccurate logging responses will be screened by unreliable gas test results for different types of fluids. It is hard to make high-precision fluid logging identification charts or models. Therefore, this article combines gas logging, well logging, testing and other data to research the test and logging data quality classification. Firstly, we select reliable standard samples through the initial gas test results. Secondly, we analyze the four main factors which affect the inaccuracy of gas test results. Thirdly, according to these factors, the flowback rate and the sand volume are determined as the main parameters. Then, we establish a recognition chart of injected sand volume/gas–water ratio. Finally, we proposed an evaluation method for testing quality classification. It provides a test basis for the subsequent identification of gas and water through the second logging interpretation. It also provides a theoretical basis for the exploration and evaluation of tight oil and gas reservoirs.

Power Quality Disturbance Identification and Optimization Based on Machine Learning

In order to improve the electrical quality disturbance recognition ability of theneural network, this paper studies a depth learning-based power quality dis-turbance recognition and classification method: constructing a power qualityperturbation model, generating training set; construct depth neural network;profit training set to depth neural network training; verify the performance ofthe depth neural network; the results show that the training set is randomlyadded 20DB-50DB noise, even in the most serious 20dB noise conditions,it can reach more than 99% identification, this is a tradition. The methodis impossible to implement. Conclusion: the deepest learning-based powerquality disturbance identification and classification method overcomes thedisadvantage of the selection steps of artificial characteristics, poor robust-ness, which is beneficial to more accurately and quickly discover the categoryof power quality issues.

Pore Structure and Movable Fluid Characteristics of Typical Sedimentary Lithofacies in a Tight Conglomerate Reservoir, Mahu Depression, Northwest China

The pore structure and movable fluid characteristics of tight conglomerate reservoirs are complex, which are greatly different from conventional reservoirs. The depositional mechanism is the fundamental factor controlling the physical properties of conglomerate reservoirs. However, there is a lack of systematic research on the pore structure and movable fluid characteristics of conglomerate reservoirs with typical sedimentary facies. This paper investigates the pore structure and movable fluid characteristics of conglomerate of different sedimentary facies based on various experiments. Casting thin sections, X-ray diffraction, scanning electron microscopy, high-pressure mercury injection, and nuclear magnetic resonance experiments were conducted on 32 conglomerates samples from the Mahu Sag, Junggar Basin, China. The quality classification method of tight conglomerate reservoirs is established. The results show that the conglomerate can be divided into three sedimentary facies; traction flow conglomerate (TFC) and pebbled sandstone (PSS) mainly develop intergranular pores and dissolved pores; and the pore diameter curves are mainly a double peak, single peak, and flat peak. Gravity flow conglomerate (GFC) mainly develops dissolved pores and interstitial micropores, and the pore diameter curve is mainly a single peak. PSS includes pebbled gritty sandstone (P(G)SS) and pebbled fine sandstone (P(F)SS). TFC and P(G)SS are favorable class I reservoirs, while GFC and P(F)SS are nonfavorable class II reservoirs. A new parameter, the ratio of the major axis to the minor axis of the pore outer ellipse (axial ratio), is proposed to quantitatively describe the compaction effect. The average axial ratios of the three lithofacies are 3.04, 3.98, and 8.78, respectively, indicating that the compaction is intensified and the pore structure becomes worse. By analyzing the correlation between pore structure parameters and permeability, it is found that the main controlling factors of permeability of GFC and TFC are sorting and connectivity, respectively, and the main flow radius is the most suitable parameter to describe permeability. A linear spectral decomposition method was used to establish a new quantitative calculation method of movable fluid saturation for different types of pores, and the results show that the movable fluid saturation of intergranular pores is the highest (average: 65.43%), and the movable fluid saturation of TFC and P(G)SS with more intergranular pores is the highest. Movable fluid saturation is inversely proportional to the content of I/S and the compaction rate and positively proportional to the content of quartz and feldspar and the cementation rate. The fluid mobility of water-wet samples is weaker. The research results provide theoretical support for the identification of favorable reservoirs and the cognition of a development mechanism.

Meta-recommendation of pork technological quality standards

Pork quality classification is supported by different reference standards that are widely reported in the literature. However, selecting the most suitable standard for each type of meat samples remains a challenge, due to their intrinsic variation according to the quality parameters’ interval. The usage of meta-learning was proposed to automatically recommend the most adequate standard for a determined sample collection, leading to a more accurate classification. The meta-learning procedure has emerged from the machine learning research field to solve the algorithm selection dilemma, outlining a new method for pork quality classification. The applicability and advantages of using a suitable classification standard for pork quality were addressed using the J48 Decision Tree (DT) algorithm, which serves as the meta-recommender. Experiments conducted with six pork standards revealed promising results based on a few meta-attributes ( L ∗, water hold capacity, and dataset entropy) as the approach successfully recommended all scenarios. • Meta-learning usage in pork quality evaluations. • L∗, WHC and dataset entropy were the most important meta-features. • We provide a comprehensive identification of porks standards. • Understanding of causes of higher incidence in poor meat quality categories.

Overview and research progress of no reference image quality evaluation methods

Image quality will greatly affect the acquisition of human visual information, but the image will inevitably cause a certain degree of distortion in the process of transmission, compression, coding and so on. At this time, the accuracy of image quality evaluation (IQA) is particularly important, which also makes IQA an important research direction in the field of image processing. IQA methods are divided into subjective image quality evaluation method and objective image quality evaluation method.there are many subjective evaluation methods, and the international telecommunications union (ITU) has also proposed many subjective evaluation methods and standards. there are three most commonly used methods, namely, the double stimulation continuous quality classification method, the single stimulation continuous quality classification method and the double stimulation damage classification method. Correspondingly, there are three kinds of objective evaluation methods, full reference, half reference and no reference. At present, this paper selects the research focus of IQA field-no reference image quality evaluation method as the main research object, summarizes the advantages and disadvantages of each non-reference evaluation method, and prospects the future development of the non-reference image quality evaluation method.

Development of Chicken Nutritional Quality Classification Methods and Algorithms Eggs Based on Characteristics of Yellow Eggs

In order to sustain human life, food is a very necessary material, since the human body requires nutrition to be used in everyday activities. In this case, eggs are among the nutritional intake, a protein that is commonly consumed by the community, particularly native chicken eggs that are available quite a lot. The nutritional quality does have beneficial advantages for the human body in all areas of the egg. In eggs, 10.8 percent of the protein content the body requires is in egg whites and 16.3 percent in egg yolks. The consistency of the egg is defined by the yolk color strength. The principal characteristic of egg quality is the yolk color. There are three basic forms of widely discovered shades of egg yolk. Pale yellow, plain yellow and orange yellow. Vitamin A, vitamin B2, folic acid, vitamin B6, and vitamin B12, iron, calcium, phosphorus, potassium, and cholesterol are rich in vitamins and minerals in egg yolks. Using the object image of 30 egg yolks, the preprocessing process with the background image is then uniform in size and then grouped by color-dependent characteristics so that the nutritional quality can be identified and the attribute extract is grouped using matlab based on sample testing at the Broad Research and Implementation of Biotechnology BPPT Testing Laboratory The convolutional process of neural networks. Development of characteristic digital image-driven extraction of chicken egg yolk and implementation of nutritional quality grouping of eggs based on characteristic colour extraction of egg yolk. In yellow colour, chicken offers 94% similarity.

Research on External Quality Automatic Detection and Classification Method of Navel Orange Based on Simple Dense Network

We propose a simple dense convolutional neural network model (SDenseNet) to solve the problem of low accuracy and poor performance in classification of navel orange with large external differences. Specifically, AlexNet is adopted as the backbone network, through the introduction of batch normalization(BN) and reduction initial size of the convolution kernel to accelerate convergence of the network model. Besides, we design a new feature reuse structure to promote the connection between layers, and modify the full connection layer by using global pooling to reduce training parameters. The experimental results demonstrate that our proposed SDenseNet significantly improves the performance with 99.33% accuracy on self-built navel orange dataset, outperforming the classic models of LeNet, AlexNet, SqueezeNet and ResNet with significant improvements of 5.33%, 1.55% and 3.55% respectively. The research results provide a new solution for external quality automatic detection and classification of navel orange.