Advanced Dielectric Anomaly Detection for Non-Destructive Fruit Inspection Using Electromagnetic Microwave Technique

Multiple studies investigating the relationship between intake of different types of fruit and colorectal cancer (CRC) risk have yielded inconsistent results. To perform a meta-analysis of existing studies to assess the association between the intake of different kinds of fruit and the incidence of CRC. We searched online literature databases including PubMed, Embase, WOS, and Cochrane Library for relevant articles available up to August 2022. With data extracted from observational studies, odds ratios (ORs) with 95% confidence intervals (CIs) were assessed using random-effects models. A funnel plot and Egger's test were used to determine publication bias. Furthermore, subgroup analysis and dose-response analysis were performed. All analyses were conducted using R (version 4.1.3). Twenty-four eligible studies involving 1068158 participants were included in this review. The meta-analysis showed that compared to a low intake, a higher intake of citrus, apples, watermelon, and kiwi reduced the risk of CRC by 9% [OR (95%CI) = 0.91 (0.85-0.97)], 25% [OR (95%CI) = 0.75 (0.66-0.85)], 26% [OR (95%CI) = 0.74 (0.58-0.94)], 13% [OR (95%CI) = 0.87 (0.78-0.96)], respectively. No significant association was observed between the intake of other types of fruit and the risk of CRC. In the dose-response analysis, a nonlinear association was found [R (95%CI) = -0.0031 (-0.0047 to -0.0014)] between citrus intake and CRC risk (P < 0.001), with the risk minimized around 120 g/d (OR = 0.85), while no significant dose-response correlation was observed after continued increase in intake. We found that a higher intake of citrus, apples, watermelon, and kiwi was negatively associated with the risk of CRC, while the intake of other types of fruits were not significantly associated with CRC. Citrus intake showed a non-linear dose-response relationship with the risk of CRC. This meta-analysis provides further evidence that a higher intake of specific types of fruit is effective in preventing the occurrence of CRC.

There are variety of fruits around the world, different types of fruits contain different types of nutrients and vitamins which could benefits our health. In order to understand which fruit can provide specific type of nutrients, we need to identify the types of fruits. However, fruits grow in a different shape, colour and texture based on the country they were planted and the environment of the land. Implementing a machine vision-based recognition on the fruits can help people recognize them easily. In this paper, an edge detection method is applied using computer vision approach to recognize different types of fruits. The fruits are classified based on the features extracted from their images. In the experiment, a total of 450 images of three types of fruit are used, which are apples, lemons and mangoes. Pre-processing steps are applied on the captured image to improve the quality of fruit details and the edge features are extracted using Canny Edge Detection method. Classification of the fruits is accomplished using two different types of learning model, the deep leaning model, Convolution Neural Network (CNN) and machine learning model, Support Vector Machines (SVM). The performance of both classifiers is compared and the model with the best performance, SVM is chosen as the model for the system. The system can achieve 86% classification accuracy with the SVM model, which is good enough for fruit recognition.

A strategy for healthy eating habits of daily fruits revisited: A metabolomics study

In recent years, the number of diseases caused by fungal pathogens has increased significantly. Many species of fungi are pathogenic for plants, causing a threat to food production and to humans, and are among the causes of chronic diseases. The aim of the study is to determine the enzyme profiles of fungi, depending on the different types of fruit with which they have contact, and to determine the differences in these profiles in relation to the substrate on which they are grown. Six strains of fungi identified as Cladosporium sphaerospermum, Fusarium poae, Alternaria alternata, Penicillium expansum, Penicillium verucosum and Acremonium strictum, isolated from fruits, were selected and analyzed for enzymatic profiles. The enzymatic activity was assessed using the API ZYM test (bioMerieux, France). In the majority of the 6 fungal strains isolated from fruits, enzymes belonging to glycol-hydrolases were the most active. The exception was Acremonium strictum, where phosphatases dominated. Among most fungal isolates, the enzymes β- glucosidase and N-acetyl-β-glucosaminidase showed the highest activity. The highest β-glucosidase activities were found in Cladosporium sphaerospermum and Penicillium expansum. On the other hand, lipase, α-fucosidase and α-chymotrypsin showed the least activity. The least activity of these enzymes or their complete absence was observed in Fusarium poae, Alternaria alternata, Penicillium expansum and Acremonium strictum. The activity of hydrolytic enzymes in the isolated fungi depended on the addition of fruit and the type of medium. Individual fruits can increase or decrease the activity of the enzymes. Fungi present in fruit have pathogenic properties and can be possible risk factors for fungal infections.

First report on organochlorine pesticides in water in a highly productive agro-industrial basin of the Central Valley, Chile

Fruit is a crucial component of daily diets, emphasizing the importance of ensuring its freshness. Our proposed method utilizes YOLOv8 and SE attention mechanism for detecting the freshness of fruits. Our method balances accuracy and real-time performance while detecting the freshness of different types of fruits. The proposed method achieves an average accuracy of 87.8% and a maximum accuracy of 95.0% for detecting the freshness level of a single fruit category. Experimental results confirm the effectiveness of our proposed method in accurately detecting and localizing the freshness level of fruits, demonstrating its potential for wide application and promotion in the fruit industry.

Health as a form of human capital is crucial in the production of market and non-market commodities. It is, therefore, imperative that individuals/households invest in health as it directly influences their capacity to be productive and generate wealth. Investment in health is achieved through use of market and non-market goods and individuals own time as inputs. This study aimed at establishing the ways through which households/individuals produce their health; effect of health knowledge on health and providing policy measures to improve the health of households. A 2SLS translog health production function was employed using primary data obtained from a sample of households in Kangundo and Matungulu Sub-Counties. The study found that, the residents had normal weight according to World Health Organization classification and hence good health status, though in every 100 residents three were obese, 31 overweight, and nine were underweight. According to the study, health status improved with the frequency of fruits and vegetables consumed by individuals/households. Moreover, moderate exercise was found to be good for improved health status though excessive exercising had deleterious effect on health capital. Health knowledge was negatively related to health status, whereas it was positively related with the quadratic term for health knowledge. The study recommends that households be encouraged to increase the intake of fruits and vegetables. This could be achieved by encouraging small-scale farmers to practice horticulture, by intercropping different types of fruits and green leafy vegetables. Further, households need to be empowered with more health knowledge, by integrating health education in school curriculum from early childhood to university levels, though; this should not be tested in the final national examinations. It would be appropriate to deploy couples in the same areas, not only to reduce HIV/AIDS incidence among married people, but also improve their health status as established by the study.

Can eating five fruit and veg a day really keep the doctor away?

Fruits as the vehicle of drug resistant pathogenic yeasts

A conceptual framework and optimization for a task-based modular harvesting manipulator

In this chapter, a solar powered pesticide-spraying RFID robot has been designed and developed for its use in different types of fruits and vegetable fields. The robot is equipped with DC motors, DC pump, a solar panel, and an ultrasonic sensor. The robot's navigation has been automated using RFID (radio frequency identification) technology. An ultrasonic sensor is used for sensing any obstacle in the path of the robot's navigation. A solar panel of 30W is used for charging the batteries that are used for driving the DC motors and the DC pump. The wheels of the robot are powered by DC motors, and the DC pump is used for spraying pesticides through the nozzles. The height of the nozzles can be varied depending on the size of the crop. The robot has been designed for spraying pesticides automatically thereby controlling pests, improving productivity, and reducing the health hazards of farmers posed by pesticides.

The quality assessment and timely detection of fruit ripeness are crucial for optimizing harvest schedules, ensuring product freshness, and minimizing post-harvest losses. This abstract presents a smart detection system designed to accurately determine fruit ripeness using non-invasive sensing techniques. To evaluate the system's performance, experiments are conducted using different types of fruits at varying ripeness stages. The results demonstrate high accuracy and reliability in fruit ripeness prediction, with minimal false positives or negatives. It eliminates the need for manual inspection, reducing labour costs and potential subjective bias. It also provides non-destructive testing, allowing fruits to remain intact for further processing or sale. Furthermore, the system can be integrated into automated sorting and packaging processes, enabling efficient sorting based on ripeness and improving overall supply chain management.

One of the key elements in predicting yield is the real-time detection of fruits and vegetables. Conventional detection techniques encounter constraints in precisely discerning distinct growth stages, primarily attributable to pronounced occlusion stemming from proximal foliage, substantial intermingling among adjacent fruits, dissimilarities in dimensions, pigmentation, aggregation density, and diverse developmental attributes. An improved YOLO-v3 model is suggested for recognising apples in orchards with varying light, complex backdrops, overlapping apples, branches, and leaves in order to detect the fruits at various development phases. Images of young, growing, and ripe apples are the first things that are gathered. Subsequently, rotation-based transformations, color equilibrium adjustments, luminance manipulations, and image blurring procedures are employed to augment the visual quality of these photographs. In recent fruit detection models, both Faster R-CNN with VGG16 net, and original YOLO-v3-dense are outperformed by an improved model YOLO-v3-dense. In a scenario including a complicated orchard, the Dense-YOLOv4 model has been used to identify several mango growth phases with a high degree of occultation. By concluding all above points, it is suggested that Dense-YOLOv4 model is one of most accurate model to detect the different types of fruits and vegetables.

The project aims to develop an intelligent system capable of classifying different types of fruits based on their images. The model utilizes Convolutional Neural Networks (CNNs) to identify and categorize fruits such as apples, bananas, oranges, and more, offering high accuracy in distinguishing between various fruit types. In addition to classification, the system also evaluates the quality of the fruits, identifying features such as ripeness, blemishes, and texture, which are crucial for determining fruit quality. The model is trained on a large dataset of fruit images, enabling it to assess both visual and physical characteristics effectively. The project combines computer vision and deep learning techniques to automate the fruit classification and quality inspection process, which can significantly benefit industries like agriculture, retail, and food processing by improving product selection, reducing waste, and enhancing consumer satisfaction.

Many applications require precise estimates of high-dimensional covariance matrices. The standard estimator is the sample covariance matrix, which is conceptually simple, fast to compute and has favorable properties in the limit of infinitely many observations. The picture changes when the dimensionality is of the same order as the number of observations. In such cases, the eigenvalues of the sample covariance matrix are highly biased, the condition number becomes large and the inversion of the matrix gets numerically unstable.&#13;\nA number of alternative estimators are superior in the high-dimensional setting, which include as subcategories structured estimators, regularized estimators and spectrum correction methods. In this thesis I contribute to all three areas. &#13;\nIn the area of structured estimation, I focus on models with low intrinsic dimensionality. I analyze the bias in Factor Analysis, the state-of-the-art factor model and propose Directional Variance Adjustment (DVA) Factor Analysis, which reduces bias and yields improved estimates of the covariance matrix.&#13;\nAnalytical shrinkage of Ledoit and Wolf (LW-Shrinkage) is the most popular regularized estimator. I contribute in three aspects: first, I provide a theoretical analysis of the behavior of LW-Shrinkage in the presence of pronounced eigendirections, a case of great practical relevance. I show that LW-Shrinkage does not perform well in this setting and propose aoc-Shrinkage which yields significant improvements. Second, I discuss the effect of autocorrelation on LW-Shrinkage and review the Sancetta-Estimator, an extension of LW-Shrinkage to autocorrelated data. I show that the Sancetta-Estimator is biased and propose a theoretically and empirically superior estimator with reduced bias. Third, I propose an extension of shrinkage to multiple shrinkage targets. Multi-Target Shrinkage is not restricted to covariance estimation and allows for many interesting applications which go beyond regularization, including transfer learning. I provide a detailed theoretical and empirical analysis.&#13;\nSpectrum correction approaches the problem of covariance estimation by improving the estimates of the eigenvalues of the sample covariance matrix. I discuss the state-of-the-art approach, Nonlinear Shrinkage, and propose a cross-validation based covariance (CVC) estimator which yields competitive performance at increased numerical stability and greatly reduced complexity and computational cost. On all data sets considered, CVC is on par or superior in comparison to the regularized and structured estimators.&#13;\nIn the last chapter, I conclude with a discussion of the advantages and disadvantages of all covariance estimators presented in this thesis and give situation-specific recommendations. In addition, the appendix contains a systematic analysis of Linear Discriminant Analysis as a model application, which sheds light on the interdependency between the generative model of the data and various covariance estimators.