Fruit Classification Research Articles

XAI-FruitNet: An explainable deep model for accurate fruit classification

In agricultural technology, precise fruit classification is essential yet challenging due to inherent interclass similarities and intra-class variabilities among fruit species. Despite their impressive performance, traditional deep learning models suffer from a lack of interpretability, which hampers their transparency and trustworthiness in practical applications. To address these issues, we present XAI-FruitNet, a novel hybrid deep learning architecture designed to enhance feature discrimination by integrating average and max pooling techniques. XAI-FruitNet, an optimized architecture for efficiency evaluated across the Fruits-360, Fruit Recognition, Fruit and Vegetables Image Recognition, and Dry Fruit datasets, consistently achieves over 97 % accuracy, surpassing existing state-of-the-art models and underscoring its remarkable generalization capability. A significant advancement of XAI-FruitNet is its built-in interpretability, which enhances the model's transparency and fosters trust among endusers. Through rigorous experimentation, we demonstrate that XAI-FruitNet advances state-of-the-art fruit classification accuracy and sets a new standard for explainable artificial intelligence (XAI) in agricultural applications. This hybrid approach ensures that stakeholders can rely on the classification outcomes' high performance and comprehensible nature, thereby offering a robust and trustworthy solution for modern agricultural needs.

The Investigation of Hyperparameters Influence on Fruit and Vegetable Image Recognition Performance Using SVM

This study aims to improve the efficiency of automatic classification and quality control of fruits and vegetables through image recognition technology, to achieve efficient and accurate intelligent sorting in agricultural production, reduce labor costs and improve market competitiveness. A high-quality image dataset of 36 fruit and vegetable categories from Kaggle is used in this study. The images in the dataset have been preprocessed to ensure that the data is suitable for the classification task and sets the stage for efficient training and evaluation of the model. Logistic regression was first used as the baseline model in order to compare the performance with the Support Vector Machine (SVM) model. Subsequently, hyperparameter tuning is performed to optimize the model to achieve the best cross-validation accuracy. Next, the SVM model is trained with the selected hyperparameters, and the training time is recorded. The performance of the model was evaluated in detail by confusion matrix and classification reports, and the test set was used for final validation to ensure that the model would also perform well on unseen data. The SVM model achieves an accuracy of 96% on both validation and test sets, which is a very good performance. The hyperparameters optimized by GridSearchCV (C = 10, gamma = 0.1, kernel = rbf) effectively improve the performance of the model, verifying that reasonable hyperparameter selection is crucial to the SVM model. These results show that the model has good generalization ability and potential to be applied to real-time classification tasks.

Machine Learning in Agriculture: KNN-Based Classification of Fruits and Vegetables

In modern agriculture and the food industry, it is essential to classify fruits and vegetables accurately and efficiently to meet growing consumer demand and reduce post-harvest losses. Traditional manual methods are often labor-intensive and error prone, highlighting the need for automated solutions. This paper discusses the application of the k Nearest Neighbor (KNN) algorithm to fruit and vegetable recognition using image processing technology. In this study, image data sets are used for feature extraction using Directional Gradient Histogram (HOG) and dimensionality reduction using Principal Component Analysis (PCA). The accuracy of KNN model on verification set and test set reaches 97%, which proves its validity. Confusion matrix analysis and F1 score evaluation further revealed the strengths and areas of improvement of the model, particularly in distinguishing visually similar categories. The results show that the integration of artificial intelligence (especially KNN) offers great potential for the automation of agricultural classification tasks. Future studies could combine more advanced models, such as CNNS, with larger datasets to improve accuracy and robustness.

Hardware Design Strategies Oriented to Edge Computing and AgriFood Electronics for Image Processing using Cellular Automata

There has been a growing interest from academia and industry in developing circuits and systems for edge computing and quality control tasks in food production lines, where image-processing is frequently required. This paper outlines the required considerations for designing a fruit classification system based on image-processing using Cellular Automata (CA) models and integrating it into reconfigurable hardware (HW) such as Field Programmable Gate Arrays (FPGAs). Parallel processing in CA requires numerous processing elements to be implemented and mapping CA models to HW generally comes with limitations. Homogeneous CA arrays are easier to design and implement in HW but can be resource-demanding. To fill this gap, this study explores different alternatives for the HW implementation of CA models, particularly trading computational-parallelism for a more optimized use of the available HW resources. We conducted experimental tests of the designed HW system using the Digilent Nexys development board, and the operation was validated against software-based benchmarks for image-processing, particularly concerning edge-detection. The presented study provides a broader range of design solutions for the HW implementation of two-dimensional CA models and a better understanding of their advantages and disadvantages. The results show that solutions focusing on instruction-parallelism add some complexity to the conception and require more design effort, compared to homogeneous CA models composed of identical cells. However, the instruction-parallel design solutions can significantly improve the HW resource utilization, especially when implementing computationally intensive CA rules in FPGAs.

Classification Of Plants By Their Fruits And Leaves Using Convolutional Neural Networks

The population growth of the world is exponential, this makes it imperative that we have an increase in food production. In this light, farmers, industries and researchers are struggling with identifying and classifying food plants. Over the years, there have been challenges that come with identifying fruits manually. It is time-consuming, labour intensive and requires experts to identify fruits because of the similarity in fruit’s leaves (citrus family), shapes, sizes and colour. A computerized detection technique is needed for the classification of fruits. Existing solutions to fruits classifications are majorly based on fruit or leave used as input. A new model using Convolutional Neural Network (CNN) is proposed for fruits classification. A dataset of 5 classes of fruits and fresh dry leaves plants (Mango, African almond, Guava, Avocado and Cashew) comprising of 1000 images each. The proposed model hyperparameters were: Conv2D layer, activation layer, dense layer, a learning and dropout rates of 0.001 and 0.5 respectively were used for the experiment. Various performances for accuracies of 91%, 97%, 78% and 97% were obtained for proposed model on local dataset, proposed model on benchmark dataset, benchmark model on local dataset and benchmark model on benchmark dataset. The proposed model is robust on both local and benchmark datasets and can be used for effective classification of plants

Issues of classification and examination of lemons in the course of forensic commodity expertise

Identification and examination of export-import goods, in particular lemons, plays a significant role in facilitating the detection of customs violations and crimes, as well as in ensuring preventive measures to eliminate them. The article analyses the peculiarities of examination of citrus fruits, in particular lemons, with a view to their classification under classifier of goods Commodity Nomenclature of Foreign Economic Activity. The methodological approaches to the classification of citrus fruits within the framework of commodity expertise are considered. The main diagnostic tasks to be solved by a forensic expert in the course of lemon examination are identified: determination of product quality, chemical analysis, phytopathological analysis, toxicological analysis, organoleptic analysis, microbiological analysis, identification of origin, assessment of storage and transportation conditions. Modern analytical methods allow to conduct a comprehensive study of lemons at various levels – molecular, chemical and morphological – to obtain objective data on their quality, safety and other important characteristics. This data can be useful in legal proceedings to determine the origin, compliance with quality standards and dispute resolution. The article summarises information on the typical defects of lemons, namely: mechanical damage, diseases and lesions, growth and development defects, skin defects, parasites and pests, physiological defects, chemical defects (these defects are of natural origin). Defects inherent in low-quality products with signs of deterioration include mechanical damage, rot and mould, discolouration, wrinkling, shape defects, pests, disease damage, additional chemical treatment, loss of aroma and taste, improper storage and transportation, black spots, and skin tears. The mechanism for conducting a forensic examination of lemons is described in detail.

Automated Mango Classification Using Convolutional Neural Networks (CNN)

This paper presents a system developed for the automated classification of different mango varieties using Convolutional Neural Networks (CNNs). The model was trained on an image dataset containing labeled mango varieties, which was augmented to enhance robustness. The CNN architecture comprises convolutional layers, pooling layers, and fully connected layers, optimized using TensorFlow. The system achieved satisfactory accuracy on both training and validation datasets. Evaluation was conducted using confusion matrices and training curves. The proposed system can classify mango images in real- time, providing predictions with confidence scores. The results demonstrate the potential of deep learning in automating fruit classification tasks, offering significant benefits for agricultural and retail sectors by improving efficiency and accuracy.

Apple fruits categorizing based on deep convolutional neural network techniques

<p>For a variety of reasons, including the high degree of similarity between varieties of the same type of fruit, the requirement to train the technique on a large amount of data, and the type and number of features suitable for application, the use of computer vision techniques in the classification of fruits still faces many challenges. Additionally, the technique's effectiveness and speed both need to be improved. Deep conventional neural network (DCNN) approaches were required for all of these reasons. A proposed CNN model is described in this work. The suggested methodology is intended to quickly and accurately categorize thirteen groups of apple fruits. The proposed technique was based on training and testing the model on a maximum number of images of apple fruits, by increasing the number of database images tenfold, after augmentation was performed on the images. The technology also relied on good tuning of the hyperparameters. To further ensure the efficiency of training, validation was performed on 20% of the database. All results that demonstrate the high efficiency of the proposed model were reviewed. The results of the proposal were compared with the results of four related techniques. The results showed the great advantage of the proposed technology at all levels.</p>

Fruit classification by assessing slice hardness based on RGB imaging. Case study: apple slices

Fruit classification by assessing slice hardness based on RGB imaging. Case study: apple slices

Bilberry metabolomic and proteomic profiling during fruit ripening reveals key dynamics affecting fruit quality.

Bilberry (Vaccinium myrtillus L.) is a wildberry species that is prevalent in northern Europe. It is renowned and well-documented for its nutritional and bioactive properties, especially due to its anthocyanin content. However, an overview of biological systems governing changes in other crucial quality traits, such as size, firmness, and flavours, has received less attention. In the present study, we investigated detailed metabolomic and proteomic profiles at four different ripening stages of bilberry to provide a comprehensive understanding of overall quality during fruit ripening. By integrating omics datasets, we revealed a novel global regulatory network of plant hormones and physiological processes occurring during bilberry ripening. Key physiological processes, such as energy and primary metabolism, strongly correlate with elevated levels of gibberellic acids, jasmonic acid, and salicylic acid in unripe fruits. In contrast, as the fruit ripened, processes including flavour formation, cell wall modification, seed storage, and secondary metabolism became more prominent, and these were associated with increased abscisic acid levels. An indication of the increase in ethylene biosynthesis was detected during bilberry development, raising questions about the classification of non-climacteric and climacteric fruits. Our findings extend the current knowledge on the physiological and biochemical processes occurring during fruit ripening, which can serve as a baseline for studies on both wild and commercially grown berry species. Furthermore, our data may facilitate the optimization of storage conditions and breeding programs, as well as the future exploration of beneficial compounds in berries for new applications in food, cosmetics, and medicines.

Computer Vision based Greenhouse Fruits and Vegetables Identification – A Review

In recent years, computer vision has emerged as a powerful tool for automating various processes in agriculture, particularly in greenhouse environments. The identification and classification of vegetables and fruits within greenhouses play a crucial role in monitoring crop growth, assessing health status, and optimizing resource utilization. This review study provides a comprehensive overview of recent advancements in computer vision techniques for the identification of vegetables and fruits in greenhouse settings. This study discusses about various methodologies and challenges in this domain, aiming to provide insights for researchers and practitioners interested in leveraging computer vision for greenhouse agriculture.

A Deep Learning Approach for Detection of Disease and Classification of Fruits

The calorie component of fruit is identified using a common photo dataset and several machine learning advancements, including Pre-processing, segmentation, function extraction, and classification which is primarily based on size and shape. Using image processing methods, the dimensions of fruit objects are calculated. Finally, we'll provide users and patients with the best advice for fruit consumption based on our best estimates of the number of calories in the fruit. To help farmers save time and money while improving the accuracy of plant disease detection, we have developed a technology. Thanks to the usage of many speed-enhancing methods, it outperforms competing systems. An easy-to- implement method for picture segmentation, classification, and reconstruction, the SVM Algorithm is what we're employing here. Even when the objective function isn't acting well, it may nevertheless perform well during global optimization.

Exploring the Application of K-means Machine Learning Algorithm in Fruit Classification

Exploring the Application of K-means Machine Learning Algorithm in Fruit Classification

DeepDate: A deep fusion model based on whale optimization and artificial neural network for Arabian date classification.

As agricultural technology continues to develop, the scale of planting and production of date fruit is increasing, which brings higher yields. However, the increasing yields also put a lot of pressure on the classification step afterward. Image recognition based on deep learning algorithms can help to identify and classify the date fruit species, even in natural light. In this paper, a deep fusion model based on whale optimization and an artificial neural network for Arabian date classification is proposed. The dataset used in this study includes five classes of date fruit images (Barhi, Khalas, Meneifi, Naboot Saif, Sullaj). The process of designing each model can be divided into three phases. The first phase is feature extraction. The second phase is feature selection. The third phase is the training and testing phase. Finally, the best-performing model was selected and compared with the currently established models (Alexnet, Squeezenet, Googlenet, Resnet50). The experimental results show that, after trying different combinations of optimization algorithms and classifiers, the highest test accuracy achieved by DeepDate was 95.9%. It takes less time to achieve a balance between classification accuracy and time consumption. In addition, the performance of DeepDate is better than that of many deep transfer learning models such as Alexnet, Squeezenet, Googlenet, VGG-19, NasNet, and Inception-V3. The proposed DeepDate improves the accuracy and efficiency of classifying date fruits and achieves better results in classification metrics such as accuracy and F1. DeepDate provides a promising classification solution for date fruit classification with higher accuracy. To further advance the industry, it is recommended that stakeholders invest in technology transfer programs to bring advanced image recognition and AI tools to smaller producers, enhancing sustainability and productivity across the sector. Collaborations between agricultural technologists and growers could also foster more tailored solutions that address specific regional challenges in date fruit production.

Interpretation and explanation of computer vision classification of carambola (Averrhoa carambola L.) according to maturity stage

The classification of carambola, also known as starfruit, according to quality parameters is usually conducted by trained human evaluators through visual inspections. This is a costly and subjective method that can generate high variability in results. As an alternative, computer vision systems (CVS) combined with deep learning (DCVS) techniques have been introduced in the industry as a powerful and an innovative tool for the rapid and non-invasive classification of fruits. However, validating the learning capability and trustworthiness of a DL model, aka black box, to obtain insights can be challenging. To reduce this gap, we propose an integrated eXplainable Artificial Intelligence (XAI) method for the classification of carambolas at different maturity stages. We compared two Residual Neural Networks (ResNet) and Visual Transformers (ViT) to identify the image regions that are enhanced by a Random Forest (RF) model, with the aim of providing more detailed information at the feature level for classifying the maturity stage. Changes in fruit colour and physicochemical data throughout the maturity stages were analysed, and the influence of these parameters on the maturity stages was evaluated using the Gradient-weighted Class Activation Mapping (Grad-CAM), the Attention Maps using RF importance. The proposed approach provides a visualization and description of the most important regions that led to the model decision, in wide visualization follows the models an importance features from RF. Our approach has promising potential for standardized and rapid carambolas classification, achieving 91 % accuracy with ResNet and 95 % with ViT, with potential application for other fruits.

Deep Learning for Automatic Classification of Fruits and Vegetables: Evaluation from the Perspectives of Efficiency and Accuracy

Within the agricultural domain, accurately categorizing the freshness levels of fruits and vegetables holds immense significance, as this classification enables early detection of spoilage and allows for appropriate grouping of products based on their intended export destinations. These processes necessitate a system capable of meticulously classifying fruits and vegetables while minimizing labor expenditures. The current study concentrates on developing an advanced model that can effectively categorize the freshness status of each fruit and vegetable as 'good,' 'medium,' or 'spoiled.' To achieve this objective, various artificial intelligence models, including CNN, AlexNet, ResNet50, GoogleNet, VGG16, and EfficientB3, have been implemented, attaining remarkable success rates of 99.75%, 97.97%, 96.71%, 99.49%, 98.75%, and 99.81%, respectively.

Application of K-NN algorithm using gray level co-occurrence matrix for mango fruit classification cased on leaf image

Application of K-NN algorithm using gray level co-occurrence matrix for mango fruit classification cased on leaf image

Peculiarities of the examination of citrus fruits (mandarins) during the forensic examination

In today’s environment, the issue of ensuring the identification and safety of imported products is of particular relevance, which is aimed at the successful functioning of the state in the global economy and the reduction of technical barriers to trade. The article discusses the peculiarities of conducting an examination of citrus fruits, in particular, tangerines, with a view to their further classification in the Ukrainian Classifier of Goods for Foreign Economic Activity. Approaches to the classification of citrus fruits in the process of commodity examination are summarized. The diagnostic tasks to be solved by the forensic expert during the study of citrus fruits are defined, which will allow to have more analytical information from this category of goods. The main identification tasks of the commodity examination of citrus fruits, in particular mandarins, are: determination of the variety or species of mandarins, detection of certain quality markers, determination of origin, assessment of the degree of freshness and maturity, detection of deviations from standards. Modern methods of analysis are indicated, which allow to carry out research on tangerines at the molecular, chemical and morphological levels in order to obtain objective information about their quality, safety and other characteristics that may be important in court proceedings. The main aspects of microbiological methods during the examination include: detection of pathogenic microorganisms (analysis for the presence of pathogenic bacteria such as Salmonella, E. coli and Listeria monocytogenes), determination of the level of the total microbial flora, assessment of the degree of rot and mold, determination of the effectiveness of preservatives and other preservation methods, assessment of compliance with food safety standards. Systematization of information data on characteristic defects of citrus fruits (tangerines), namely: rot, mold, mechanical damage, weak skin color, increased acidity, defects in shape or size (identified defects are of natural origin) is presented. Defects that are characteristic of low-quality products with signs of deterioration: rot, mold, mechanical damage, signs of drying or desiccation, heterogeneity in color and texture. The mechanism of forensic commodity examination of citrus fruits (tangerines) is described.

Meta-Analisis Pengaruh Blansir terhadap Total Fenolik Buah dan Sayur

Blanching is the most common pretreatment method for fruits and vegetables to maintain product quality. Numerous studies revealed desirable impacts of the process to total phenolic content, while the others showed opposite results. In addition, some differences in the blanching condition such as methods and classification of fruits and vegetables were also known able to affect total phenolic content. This study aimed to analyze the effect of blanching on total phenolic content in vegetables and fruits through a meta-analysis approach. Study selection was conducted with a PRISMA method. There were 28 articles and 82 data used for meta-analysis. The data were analyzed by Hedges’d to calculate the effect size with confidence interval (CI) utilizing OpenMee software. The result showed that blanching demonstrated significant effects on the rise of total phenolic content (hedges’d[95% CI]: 0.782 [0.037 to 1.527]; p<0.05). Based on fruits and vegetables subgroups, the blanching significantly affected the increase of total phenolic content in leaves (hedges’d[95% CI]: 1.484 [0.185–2.782]) and fruits (hedges’d[95% CI]: 2.504 [0.153–4.854] compared to stems and stalks, fruits as vegetable, shoots and sprouts, pulse, and flowers. Meanwhile, in terms of blanching method subgroups, hot water blanching (hedges’d[95% CI]: 1.096 [0.234–1.959]) had significant effects on the rise of total phenolic content compared to steam blanching. Moreover, this meta-analysis was robust against publication bias.

Research Status and Development Trend of Key Technologies for Pineapple Harvesting Equipment: A Review

Pineapple harvesting is a key step in pineapple field production. At present, pineapple fruits are usually picked manually. With decreasing labor resources and increasing production costs, machines have been used instead of manual picking approaches in the modern pineapple industry. This paper briefly describes the basic situation of pineapple planting worldwide. Based on the degree of automation of mechanized pineapple harvesting equipment, the main structural forms, core technologies, and operation modes of semi-automatic, automatic, and intelligent pineapple harvesting equipment are summarized. The research status and existing problems of key pineapple fruit picking robots, such as fruit recognition, maturity classification, positioning, and separation of pineapple fruits, are analyzed. Considering the problems of pineapple harvesting equipment, such as difficulty entering the ground, low harvesting efficiency, low picking success rate, and fruit damage, innovative future research directions for mechanized pineapple harvesting technology are proposed, such as combining agricultural machinery and agronomical principles, integrating mechanized, automated, and intelligent technology, and developing modular designs and generalized approaches.