| Titre : | Detection and identification of original honey using AI |
| Auteurs : | Anfal Lastab, Auteur ; Fatiha Elgarni, Auteur ; Adel Abdelli, Directeur de thèse |
| Type de document : | Mémoire magistere |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2025 |
| Format : | 1 vol. (77 p.) / ill.couv.ill.encoul / 30cm |
| Langues: | Anglais |
| Langues originales: | Anglais |
| Résumé : |
This project aims to develop a reliable and efficient system for authenticating honey and determining its botanical origin. The system identifies the floral sources from which the nectar was collected by analyzing the pollen content in the honey. To achieve this, advanced analytical methods are employed to examine the pollen, helping to detect possible adulteration and ensure that the honey is not mixed with artificial substances or lower-quality alternatives.By identifying the botanical sources, the system also enhances product traceability,providing essential information about the honey’s geographical and floral origin. Thisauthentication process not only maintains the integrity of honey but also plays a crucialrole in protecting consumers from fraudulent products. By offering verified informationabout the honey’s origin and quality, the system boosts consumer confidence and enables laboratories to differentiate authentic honey from adulterated products.Ultimately, this project contributes to strengthening trust among honey producers,regulatory authorities, and consumers by ensuring that honey available in the marketmeets established quality standards and remains free from contamination or misleadingclaims.Our system assists laboratories in classifying pollen by automating this tedious and time-consuming task. It was built using convolutional neural networks (CNNs), a deep learning method known for its effectiveness in image classification. The system achievedan accuracy of 91% in classifying 23 different types of pollen. This AI model was integratedinto a mobile application, which can later be connected to a microscope to assist laboratories in classifying honey pollen samples. |
| Sommaire : |
General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Chapter 1: Background 26 1.1 Definition of Honey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.2 Importance of Honey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.2.1 Nutritional Value . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.2.2 Medicinal and Therapeutic Benefits . . . . . . . . . . . . . . . . . 28 1.2.3 Economic and Environmental Importance . . . . . . . . . . . . . 28 1.2.4 Cultural and Religious Significance . . . . . . . . . . . . . . . . . 28 1.3 The issue of Honey Adulteration . . . . . . . . . . . . . . . . . . . . . . . 29 1.3.1 Definition of Honey Adulteration . . . . . . . . . . . . . . . . . . 29 1.3.2 Methods of Honey Adulteration . . . . . . . . . . . . . . . . . . . 29 1.3.3 Impact of Honey Adulteration . . . . . . . . . . . . . . . . . . . . 30 1.4 Challenges in Detecting Natural vs. Adulterated Honey . . . . . . . . . . 30 1.4.1 Variability in Natural Honey Composition . . . . . . . . . . . . . 30 1.4.2 Advanced Techniques Used in Honey Modification . . . . . . . . 31 1.4.3 Limitations of Traditional Detection Methods . . . . . . . . . . . 31 1.4.4 Inconsistent Honey Regulations and Standards . . . . . . . . . . . 31 1.4.5 High Costs and Accessibility of Advanced Testing . . . . . . . . . 32 1.4.6 Consumer Awareness and Market Challenges . . . . . . . . . . . . 32 1.5 Limitations of Traditional Honey Authentication . . . . . . . . . . . . . . 32 1.5.1 Physical Tests Are Inaccurate and Unreliable . . . . . . . . . . . 32 1.5.2 Limitations of Chemical Analysis Methods . . . . . . . . . . . . . 33 1.5.3 Microscopic and Pollen Analysis Challenges . . . . . . . . . . . . 35 1.5.4 Regulatory and Standardization Issues . . . . . . . . . . . . . . . 36 1.5.5 Economic and Accessibility Challenges . . . . . . . . . . . . . . . 37 1.6 Palynology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 1.6.1 Pollen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 1.6.2 Composition of a pollen grain . . . . . . . . . . . . . . . . . . . . 38 1.6.3 Structure of a Pollen Grain . . . . . . . . . . . . . . . . . . . . . 39 1.6.4 Main types of pollen . . . . . . . . . . . . . . . . . . . . . . . . . 39 1.6.5 Pollen Types and Corresponding Plants . . . . . . . . . . . . . . 41 1.6.6 Pollen Analysis in Honey Authentication . . . . . . . . . . . . . . 41 1.7 Key Features for Honey Classification . . . . . . . . . . . . . . . . . . . 42 1.7.1 Microscopy-Based Classification . . . . . . . . . . . . . . . . . . . 42 1.7.2 Spectral Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 1.8 The role of artificial intelligence in food authentication . . . . . . . . . . 45 1.8.1 Definition of Food Authentication . . . . . . . . . . . . . . . . . . 45 1.8.2 The role of artificial intelligence . . . . . . . . . . . . . . . . . . . 45 1.9 Methods Needed for an AI-Based Approach . . . . . . . . . . . . . . . . 46 1.9.1 Importance of Using AI for Detecting Honey (Natural and Adulterated) 46 1.9.2 Machine Learning for Honey Authentication . . . . . . . . . . . . 46 1.9.3 Deep Learning for Honey Authentication . . . . . . . . . . . . . . 46 1.9.4 Advantages of AI Over Traditional Methods . . . . . . . . . . . . 46 1.10 Types of Honey Analyzed Using Machine Learning . . . . . . . . . . . . 47 1.10.1 Machine Learning for Natural Honey Classification . . . . . . . . 47 1.10.2 Machine Learning for Adulterated Honey Classification . . . . . . 47 1.10.3 Machine Learning for Processed Honey Classification . . . . . . . 48 1.11 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1.11.1 Microscopy-Based Analysis . . . . . . . . . . . . . . . . . . . . . . 48 1.11.2 Spectral Data Acquisition . . . . . . . . . . . . . . . . . . . . . . 48 1.11.3 Chemical Composition Analysis . . . . . . . . . . . . . . . . . . . 49 1.11.4 Machine Learning and AI-Driven Data Utilization . . . . . . . . . 49 1.12 AI techniques used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 1.12.1 Image-Based Structural Analysis . . . . . . . . . . . . . . . . . . 49 1.12.2 Mathematical Discrimination Models . . . . . . . . . . . . . . . . 50 1.12.3 Layered Decision Frameworks . . . . . . . . . . . . . . . . . . . . 50 1.12.4 Comprehensive Data Fusion . . . . . . . . . . . . . . . . . . . . . 50 1.13 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Chapter 2: Artificial Intelligence and Machine Learning 51 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.2 Artificial Intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.3 Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.3.1 Machine Learning Techniques . . . . . . . . . . . . . . . . . . . . 52 2.3.1.1 Supervised Learning . . . . . . . . . . . . . . . . . . . . 53 2.3.1.2 Unsupervised Learning . . . . . . . . . . . . . . . . . . . 54 2.3.1.3 Reinforcement Learning . . . . . . . . . . . . . . . . . . 55 2.3.1.4 Semi-Supervised Learning . . . . . . . . . . . . . . . . . 55 2.3.2 Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 2.3.3 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 2.4 Deep learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2.4.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2.4.2 Difference Between Machine Learning and Deep Learning . . . . . 57 2.4.3 Application areas of deep learning . . . . . . . . . . . . . . . . . 58 2.4.4 Deep learning architecture . . . . . . . . . . . . . . . . . . . . . . 58 2.5 Artificial neural networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.5.1 Biological neuron . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.5.2 Artificial neuron . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 2.5.3 Artificial neural networks . . . . . . . . . . . . . . . . . . . . . . . 60 2.5.4 Activation function . . . . . . . . . . . . . . . . . . . . . . . . . . 61 2.6 Convolutional neural networks . . . . . . . . . . . . . . . . . . . . . . . . 64 2.6.1 Convolutional layer: . . . . . . . . . . . . . . . . . . . . . . . . . 64 2.6.2 Pooling layer: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 2.6.3 Fully connected layer (dense layer): . . . . . . . . . . . . . . . . . 67 2.7 Popular CNN Architectures . . . . . . . . . . . . . . . . . . . . . . . . . 68 2.7.1 VGGNet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 2.7.2 EfficientNet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.7.3 LeNet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.7.4 MobileNet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 2.7.5 ResNet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 2.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Chapter 3: System Design, Implementation and Results 72 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 3.2 Foctionalty architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 3.3 Material and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.3.1 Detailed CNN architecture . . . . . . . . . . . . . . . . . . . . . . 74 3.3.2 Data description . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 3.3.3 Data preprocessing and augmentation . . . . . . . . . . . . . . . . 76 3.3.4 Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 3.3.5 Transfer learning . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 3.4 Model training and results . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.4.1 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 3.5 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 3.6 Mobile Application Development . . . . . . . . . . . . . . . . . . . . . . 84 3.6.1 Definition Of Mobile Applications : . . . . . . . . . . . . . . . . . 84 3.6.2 Why Mobile Applications Choice ? . . . . . . . . . . . . . . . . . 85 3.6.3 The Pattern Design Used . . . . . . . . . . . . . . . . . . . . . . . 85 3.6.4 Implementation tools and languages . . . . . . . . . . . . . . . . . 87 3.6.5 Conception Of The Application . . . . . . . . . . . . . . . . . . . 95 3.7 Application graphical interfaces . . . . . . . . . . . . . . . . . . . . . . . 98 3.7.1 Splash Screen : . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3.7.2 Home Page : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3.7.3 Sing up : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 3.7.4 Login Page : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 3.7.5 CameraPage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.7.6 Analyzing Result Page . . . . . . . . . . . . . . . . . . . . . . . . 108 3.7.7 History Page : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3.7.8 Chat Page : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3.7.9 Profile Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 3.7.10 Tokens Page : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 3.7.11 Admin : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 3.8 Conclusion: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 General Conclusion 119 |
| Type de document : | Mémoire master |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| MINF/955 | Mémoire master | bibliothèque sciences exactes | Consultable |
