A local binary pattern based texture descriptors for classification of tea leaves

Machine Learning-Based Detection of Dengue from Blood Smear Images Utilizing Platelet and Lymphocyte Characteristics

Dengue fever, also known as break-bone fever, can be life-threatening. Caused by DENV, an RNA virus from the Flaviviridae family, dengue is currently a globally important public health problem. The clinical methods available for dengue diagnosis require skilled supervision. They are manual, time-consuming, labor-intensive, and not affordable to common people. This paper describes a method that can support clinicians during dengue diagnosis. It is proposed to automate the peripheral blood smear (PBS) examination using Artificial Intelligence (AI) to aid dengue diagnosis. Nowadays, AI, especially Machine Learning (ML), is increasingly being explored for successful analyses in the biomedical field. Digital pathology coupled with AI holds great potential in developing healthcare services. The automation system developed incorporates a blob detection method to detect platelets and thrombocytopenia from the PBS images. The results achieved are clinically acceptable. Moreover, an ML-based technique is proposed to detect dengue from the images of PBS based on the lymphocyte nucleus. Ten features are extracted, including six morphological and four Gray Level Spatial Dependance Matrix (GLSDM) features, out of the lymphocyte nucleus of normal and dengue cases. Features are then subjected to various popular supervised classifiers built using a ten-fold cross-validation policy for automated dengue detection. Among all the classifiers, the best performance was achieved by Support Vector Machine (SVM) and Decision Tree (DT), each with an accuracy of 93.62%. Furthermore, 1000 deep features extracted using pre-trained MobileNetV2 and 177 textural features extracted using Local binary pattern (LBP) from the lymphocyte nucleus are subjected to feature selection. The ReliefF selected 100 most significant features are then fed to the classifiers. The best performance was attained using an SVM classifier with 95.74% accuracy. With the obtained results, it is evident that this proposed approach can efficiently contribute as an adjuvant tool for diagnosing dengue from the digital microscopic images of PBS.

Influence of Tea Brewing Parameters on the Antioxidant Potential of Infusions and Extracts Depending on the Degree of Processing of the Leaves of Camellia sinensis

The polyphenol content of tea depends on the growing region, harvest date, the production process used, and the brewing parameters. In this study, research was undertaken that included an analysis of the influence of the brewing process parameters on the content of total polyphenols (Folin-Ciocalteu), epigallocatechin gallate (HPLC), and antioxidant activity (against DPPH radicals) of fresh tea shrub leaves grown from Taiwan and of teas obtained from them (oolong, green in bags, and green loose from the spring and autumn harvest). The antioxidant potential was determined in the methanol and aqueous extracts, as well as in infusions that were obtained by using water at 65 or 100 °C and infusing the tea for 5 or 10 min. The highest content of total polyphenols and epigallocatechin gallate was found in green tea extracts from the spring harvest. However, in the case of infusions, the highest content of these compounds was found in green tea in bags. Steaming at 100 °C for 10 min, turned out to be the most favourable condition for the extraction. Oolong tea, brewed at 100 °C for 5 min was characterised by the highest antioxidant activity against stable DPPH radicals.

Review of Weed Detection Methods Based on Computer Vision

Weeds are one of the most important factors affecting agricultural production. The waste and pollution of farmland ecological environment caused by full-coverage chemical herbicide spraying are becoming increasingly evident. With the continuous improvement in the agricultural production level, accurately distinguishing crops from weeds and achieving precise spraying only for weeds are important. However, precise spraying depends on accurately identifying and locating weeds and crops. In recent years, some scholars have used various computer vision methods to achieve this purpose. This review elaborates the two aspects of using traditional image-processing methods and deep learning-based methods to solve weed detection problems. It provides an overview of various methods for weed detection in recent years, analyzes the advantages and disadvantages of existing methods, and introduces several related plant leaves, weed datasets, and weeding machinery. Lastly, the problems and difficulties of the existing weed detection methods are analyzed, and the development trend of future research is prospected.

Weed and Corn Seedling Detection in Field Based on Multi Feature Fusion and Support Vector Machine

Detection of weeds and crops is the key step for precision spraying using the spraying herbicide robot and precise fertilization for the agriculture machine in the field. On the basis of k-mean clustering image segmentation using color information and connected region analysis, a method combining multi feature fusion and support vector machine (SVM) was proposed to identify and detect the position of corn seedlings and weeds, to reduce the harm of weeds on corn growth, and to achieve accurate fertilization, thereby realizing precise weeding or fertilizing. First, the image dataset for weed and corn seedling classification in the corn seedling stage was established. Second, many different features of corn seedlings and weeds were extracted, and dimensionality was reduced by principal component analysis, including the histogram of oriented gradient feature, rotation invariant local binary pattern (LBP) feature, Hu invariant moment feature, Gabor feature, gray level co-occurrence matrix, and gray level-gradient co-occurrence matrix. Then, the classifier training based on SVM was conducted to obtain the recognition model for corn seedlings and weeds. The comprehensive recognition performance of single feature or different fusion strategies for six features is compared and analyzed, and the optimal feature fusion strategy is obtained. Finally, by utilizing the actual corn seedling field images, the proposed weed and corn seedling detection method effect was tested. LAB color space and K-means clustering were used to achieve image segmentation. Connected component analysis was adopted to remove small objects. The previously trained recognition model was utilized to identify and label each connected region to identify and detect weeds and corn seedlings. The experimental results showed that the fusion feature combination of rotation invariant LBP feature and gray level-gradient co-occurrence matrix based on SVM classifier obtained the highest classification accuracy and accurately detected all kinds of weeds and corn seedlings. It provided information on weed and crop positions to the spraying herbicide robot for accurate spraying or to the precise fertilization machine for accurate fertilizing.

Nondestructive monitoring of polyphenols and caffeine during green tea processing using Vis-NIR spectroscopy

Increasing consumption of green tea is attributed to the beneficial effects of its constituents, especially polyphenols, on human health, which can be varied during leaf processing. Processing technology has the most important effect on green tea quality. This study investigated the system dynamics of eight catechins, gallic acid, and caffeine in the processing of two varieties of tea, from fresh leaves to finished tea. It was found that complex biochemical changes can occur through hydrolysis under different humidity and heating conditions during the tea processing. This process had a significant effect on catechin composition in the finished tea. The potential application of visible and near-infrared (Vis-NIR) spectroscopy for fast monitoring polyphenol and caffeine contents in tea leaves during the processing procedure has been investigated. It was found that a combination of PCA (principal component analysis) and Vis-NIR spectroscopy can successfully classify the two varieties of tea samples and the five tea processing procedures, while quantitative determination of the constituents was realized by combined regression analysis and Vis-NIR spectra. Furthermore, successive projections algorithm (SPA) was proposed to extract and optimize spectral variables that reflected the molecular characteristics of the constituents for the development of determination models. Modeling results showed that the models had good predictability and robustness based on the extracted spectral characteristics. The coefficients of determination for all calibration sets and prediction sets were higher than 0.862 and 0.834, respectively, which indicated high capability of Vis-NIR spectroscopy for the determination of the constituents during the leaf processing. Meanwhile, this analytical method could quickly monitor quality characteristics and provide feedback for real-time controlling of tea processing machines. Furthermore, the study on complex biochemical changes that occurred during the tea processing would provide a theoretical basis for improving the content of quality components and effective controlling processes.

Leaf Image Recognition Based on Bag of Features

Plants are ubiquitous in human life. Recognizing an unknown plant by its leaf image quickly is a very interesting and challenging research. With the development of image processing and pattern recognition, plant recognition based on image processing has become possible. Bag of features (BOF) is one of the most powerful models for classification, which has been used for many projects and studies. Dual-output pulse-coupled neural network (DPCNN) has shown a good ability for texture features in image processing such as image segmentation. In this paper, a method based on BOF and DPCNN (BOF_DP) is proposed for leaf classification. BOF_DP achieved satisfactory results in many leaf image datasets. As it is hard to get a satisfactory effect on the large dataset by a single feature, a method (BOF_SC) improved from bag of contour fragments is used for shape feature extraction. BOF_DP and LDA (linear discriminant analysis) algorithms are, respectively, employed for textual feature extraction and reducing the feature dimensionality. Finally, both features are used for classification by a linear support vector machine (SVM), and the proposed method obtained higher accuracy on several typical leaf datasets than existing methods.

Application of Color Featuring and Deep Learning in Maize Plant Detection

Maize plant detection was conducted in this study with the goals of target fertilization and reduction of fertilization waste in weed spots and gaps between maize plants. The methods used included two types of color featuring and deep learning (DL). The four color indices used were excess green (ExG), excess red (ExR), ExG minus ExR, and the hue value from the HSV (hue, saturation, and value) color space, while the DL methods used were YOLOv3 and YOLOv3_tiny. For practical application, this study focused on performance comparison in detection accuracy, robustness to complex field conditions, and detection speed. Detection accuracy was evaluated by the resulting images, which were divided into three categories: true positive, false positive, and false negative. The robustness evaluation was performed by comparing the average intersection over union of each detection method across different sub–datasets—namely original subset, blur processing subset, increased brightness subset, and reduced brightness subset. The detection speed was evaluated by the indicator of frames per second. Results demonstrated that the DL methods outperformed the color index–based methods in detection accuracy and robustness to complex conditions, while they were inferior to color feature–based methods in detection speed. This research shows the application potential of deep learning technology in maize plant detection. Future efforts are needed to improve the detection speed for practical applications.

Effects of withering on the main physical properties of withered tea leaves and the sensory quality of congou black tea

To explore the relationship between the moisture content of withered tea leaves and their physical properties (i.e., elasticity, plasticity, flexibility, and texture) during withering, texture analyzer was employed to test the elasticity and flexibility of withered tea leaves with different moisture contents. The texture was evaluated by computer vision technology. The withered tea leaves with different moisture contents were used to process congou black tea, which was then subjected to sensory evaluation. Results showed that good elasticity, optimal flexibility, and plasticity were achieved when the moisture content of the withered tea leaves of Fudingdabai comprising two leaves and one bud varied arranging from 65.51 to 61.48%. The sensory evaluation of congou black tea revealed that moderate withering was better than long-term withering and that both moderate and long-term withering were better than no withering during processing. The moisture content was significantly correlated with the flexibility and plasticity of the withered tea leaves. Fresh tea leaves undergoing moderate withering with moisture content of 65.51-61.48% to process congou black tea, good tea shape and liquor color were achieved. This study provided new evidence that the moisture content of withered tea leaves significantly affected the quality of black tea.

Local Dominant Directional Symmetrical Coding Patterns for Facial Expression Recognition

To overcome the shortcomings of inaccurate textural direction representation and high-computational complexity of Local Binary Patterns (LBPs), we propose a novel feature descriptor named as Local Dominant Directional Symmetrical Coding Patterns (LDDSCPs). Inspired by the directional sensitivity of human visual system, we partition eight convolution masks into two symmetrical groups according to their directions and adopt these two groups to compute the convolution values of each pixel. Then, we encode the dominant direction information of facial expression texture by comparing each pixel's convolution values with the average strength of its belonging group and obtain LDDSCP-1 and LDDSCP-2 codes, respectively. At last, in view of the symmetry of two groups of direction masks, we stack these corresponding histograms of LDDSCP-1 and LDDSCP-2 codes into the ultimate LDDSCP feature vector which has effects on the more precise facial feature description and the lower computational complexity. Experimental results on the JAFFE and Cohn-Kanade databases demonstrate that the proposed LDDSCP feature descriptor compared with LBP, Gabor, and other traditional operators achieves superior performance in recognition rate and computational complexity. Furthermore, it is also no less inferior to some state-of-the-art local descriptors like as LDP, LDNP, es-LBP, and GDP.

Prediction of Moisture Content for Congou Black Tea Withering Leaves Using Image Features and Nonlinear Method

Withering is the first step in the processing of congou black tea. With respect to the deficiency of traditional water content detection methods, a machine vision based NDT (Non Destructive Testing) method was established to detect the moisture content of withered leaves. First, according to the time sequences using computer visual system collected visible light images of tea leaf surfaces, and color and texture characteristics are extracted through the spatial changes of colors. Then quantitative prediction models for moisture content detection of withered tea leaves was established through linear PLS (Partial Least Squares) and non-linear SVM (Support Vector Machine). The results showed correlation coefficients higher than 0.8 between the water contents and green component mean value (G), lightness component mean value (L^*) and uniformity (U), which means that the extracted characteristics have great potential to predict the water contents. The performance parameters as correlation coefficient of prediction set (Rp), root-mean-square error of prediction (RMSEP), and relative standard deviation (RPD) of the SVM prediction model are 0.9314, 0.0411 and 1.8004, respectively. The non-linear modeling method can better describe the quantitative analytical relations between the image and water content. With superior generalization and robustness, the method would provide a new train of thought and theoretical basis for the online water content monitoring technology of automated production of black tea.

Comparative analysis of image classification methods for automatic diagnosis of ophthalmic images

There are many image classification methods, but it remains unclear which methods are most helpful for analyzing and intelligently identifying ophthalmic images. We select representative slit-lamp images which show the complexity of ocular images as research material to compare image classification algorithms for diagnosing ophthalmic diseases. To facilitate this study, some feature extraction algorithms and classifiers are combined to automatic diagnose pediatric cataract with same dataset and then their performance are compared using multiple criteria. This comparative study reveals the general characteristics of the existing methods for automatic identification of ophthalmic images and provides new insights into the strengths and shortcomings of these methods. The relevant methods (local binary pattern +SVMs, wavelet transformation +SVMs) which achieve an average accuracy of 87% and can be adopted in specific situations to aid doctors in preliminarily disease screening. Furthermore, some methods requiring fewer computational resources and less time could be applied in remote places or mobile devices to assist individuals in understanding the condition of their body. In addition, it would be helpful to accelerate the development of innovative approaches and to apply these methods to assist doctors in diagnosing ophthalmic disease.

Automatic visual inspection of a missing split pin in the China railway high-speed

The split pin (SP) on the caliper brake is a vital component of the brake system of a bogie traveling along the China railway high-speed (CRH), and the absence of the SP could cause serious train accidents. A new automatic visual inspection method is proposed for the quick and accurate detection of SP faults of the CRH. The proposed approach is based on the histogram of gradient (HOG) combined with the complete local binary pattern (CLBP). First, a fast pyramid template matching technique is presented for localizing the region of interest to reduce the searching scope. Under the multiresolution pyramid model for target localization, a coarse-to-fine strategy is employed to ensure that the recognizing speed of the SP for the entire image is increased significantly. Second, a hierarchical framework is adopted at the localizing and inspecting stages of the SP to automatically implement the inspection tasks. To increase the robustness to the outside complex illumination, the HOG feature for localizing the target and the CLBP feature for examining the state of the SP (i.e., missing or not-missing) are extracted in the Sobel gradient domain. The localization and recognition stages are both fulfilled through the use of their respective intersection kernel support vector machine classifiers and corresponding features. In conclusion, experimental results indicate that the inspection system achieves a high accuracy rate of more than 99.0% and a real-time speed, thus proving that the proposed method is effective for the fault inspection of the SP and can satisfy the requirements of CRH's actual application.

Experimental variability and data pre-processing as factors affecting the discrimination power of some chemometric approaches (PCA, CA and a new algorithm based on linear regression) applied to (+/-)ESI/MS and RPLC/UV data: Application on green tea extracts

The influence of the experimental variability (instrumental repeatability, instrumental intermediate precision and sample preparation variability) and data pre-processing (normalization, peak alignment, background subtraction) on the discrimination power of multivariate data analysis methods (Principal Component Analysis -PCA- and Cluster Analysis -CA-) as well as a new algorithm based on linear regression was studied. Data used in the study were obtained through positive or negative ion monitoring electrospray mass spectrometry (+/-ESI/MS) and reversed phase liquid chromatography/UV spectrometric detection (RPLC/UV) applied to green tea extracts. Extractions in ethanol and heated water infusion were used as sample preparation procedures. The multivariate methods were directly applied to mass spectra and chromatograms, involving strictly a holistic comparison of shapes, without assignment of any structural identity to compounds. An alternative data interpretation based on linear regression analysis mutually applied to data series is also discussed. Slopes, intercepts and correlation coefficients produced by the linear regression analysis applied on pairs of very large experimental data series successfully retain information resulting from high frequency instrumental acquisition rates, obviously better defining the profiles being compared. Consequently, each type of sample or comparison between samples produces in the Cartesian space an ellipsoidal volume defined by the normal variation intervals of the slope, intercept and correlation coefficient. Distances between volumes graphically illustrates (dis)similarities between compared data. The instrumental intermediate precision had the major effect on the discrimination power of the multivariate data analysis methods. Mass spectra produced through ionization from liquid state in atmospheric pressure conditions of bulk complex mixtures resulting from extracted materials of natural origins provided an excellent data basis for multivariate analysis methods, equivalent to data resulting from chromatographic separations. The alternative evaluation of very large data series based on linear regression analysis produced information equivalent to results obtained through application of PCA an CA.