topic *******name********* IT******* BSc (Hons) in Information Technology Specializing in ********* Department of Information Technology Sri Lanka Institute of Information Technology Sri Lanka month year topic ******Name***** IT***** Dissertation submitted in partial fulfillment of the requirements for the Bachelor of Science Special (Honors) in Information Technology Specializing in ***** ***** Department of Information Technology Sri Lanka Institute of Information Technology Sri Lanka month year DECLARATION I declare that this is my work. This proposal does not incorporate without acknowledgement any material previously submitted for a degree or diploma in any other university or institute of higher learning. To the best of our knowledge and belief, it does not contain any material previously published or written by another person except where the acknowledgement is made in the text. Name Student ID Signature Signature: Date: Signature of the Supervisor: Date: i ABSTRACT This research is to provide a smart mobile-based solution for identifying diseases and showing possible remedies to control the diseases of selected spices plants, namely cinnamon, pepper, and cloves. Proper identification of these plants is crucial for their effective disease management. The proposed solution involves inputting an image of the plant, leaf, root, branch, or trunk to the application for identification. To achieve this main objective, specific objectives have been set out. Firstly, the visual symptoms of leaves, plant, root, branches, or trunk of spice plants will be studied and understood. This will aid in identifying the diseases affecting the plants. Secondly, the methods of capturing images on spice plants to train the model will be studied and understood. This will ensure that accurate images are used to train the model, which will enhance its accuracy. Thirdly, different techniques related to image processing, deep learning, and machine learning will be studied and understood. These techniques will be used to train the model to identify the diseases accurately. Finally, the study aims to understand how to implement a mobile application to identify diseases and show possible remedies to control the diseases. The application will be user-friendly, and the results will be displayed in a format that is easy to comprehend and use. This study seeks to provide a smart mobile-based solution for identifying diseases and showing possible remedies to control the diseases of selected spices plants. It aims to achieve this by studying and understanding the visual symptoms of the plants, how to capture images to train the model, the techniques related to image processing, deep learning, and machine learning, and how to implement a user-friendly mobile application. By providing an effective solution, the study will contribute to the proper management of the diseases of cinnamon, pepper, and cloves. Keywords : spices, identification, mobile application, disease management, deep learning ii ACKNOWLEDGEMENT iii TABLE OF CONTENT DECLARATION ........................................................................................................................... i ABSTRACT..................................................................................................................................ii ACKNOWLEDGEMENT ..............................................................................................................iii TABLE OF CONTENT ................................................................................................................. iv LIST OF TABLES ......................................................................................................................... vi LIST OF FIGURES ....................................................................................................................... vi LIST OF ABBREVIATIONS ......................................................................................................... vii LIST OF APPENDICES ............................................................................................................... vii 1 2 INTRODUCTION ................................................................................................................ 1 1.1 BACKGROUND & LITERATURE REVIEW .................................................................... 5 1.2 RESEARCH GAP ......................................................................................................... 6 1.3 RESEARCH PROBLEM................................................................................................ 7 1.4 RESEARCH OBJECTIVES ............................................................................................ 8 1.4.1 MAIN OBJECTIVE .............................................................................................. 8 1.4.2 SPECIFIC OBJECTIVES........................................................................................ 8 METHODOLOGY ............................................................................................................. 10 2.1 System Architecture ............................................................................................... 10 2.2 Development Process ............................................................................................ 11 2.2.1 2.3 Component ............................................................................................................ 11 2.4 COMMERCIALIZATION ASPECTS OF THE PRODUCT ............................................... 16 2.5 TESTING & IMPLEMENTATION ............................................................................... 17 2.5.1 3 Tools and technologies .................................................................................. 13 Testing ............................................................................................................ 17 RESULTS & DISCUSSION ................................................................................................. 20 3.1 RESULTS.................................................................................................................. 20 3.2 RESEARCH FINDINGS .............................................................................................. 21 3.3 DISCUSSION............................................................................................................ 22 4 SUMMARY OF EACH STUDENT’S CONTRIBUTION.......................................................... 23 5 CONCLUSION .................................................................................................................. 24 6 REFERENCES ................................................................................................................... 25 7 APPENDICES ................................................................................................................... 26 iv v LIST OF TABLES Table 2.1 TEST CASES ........................................................................................................ 19 LIST OF FIGURES Figure 1.1 Brain Identify what is the shape ................... Ошибка! Закладка не определена. Figure 1.2 Google Quick Draw ...................................... Ошибка! Закладка не определена. Figure 1.3 Let'sdraw.it ................................................... Ошибка! Закладка не определена. Figure 1.4 Designing Drawing Apps for Youngsters: Artistic and Technological Factors - UI ....................................................................................... Ошибка! Закладка не определена. Figure 1.5 HOW TO DRAW APP UI............................ Ошибка! Закладка не определена. Figure 2.1 System Overview Diagram ............................ Ошибка! Закладка не определена. vi LIST OF ABBREVIATIONS Abbreviations Description LIST OF APPENDICES vii 1 INTRODUCTION Spice plants such as cinnamon, pepper, and cloves can be susceptible to various diseases that can have a negative impact on their growth and yield. Understanding the types of diseases that these plants can develop and how they affect them is crucial for proper disease management. One of the common diseases that affect pepper plants is Mosaic Virus, which is caused by a virus that can affect more than 150 types of plants including fruits, vegetables, and flowers. This disease is mainly spread by insects, specifically aphids and leafhoppers, and can cause yellowing, distortion, and mottling of leaves, as well as stunted growth and reduced yields [1]. Another disease that can affect spice plants is bacterial canker, which is characterized by a thick, liquid exudate that is primarily composed of bacteria. This disease can cause cankers on the plant tissue, which can lead to the death of the affected plant. The symptoms of plant disease, such as changes in color, shape, or function of the plant, can help identify the type of disease affecting the plant [2]. Infectious plant diseases can also be caused by fungi or viruses and can range in severity from mild leaf or fruit damage to death. For example, cinnamon can be affected by powdery mildew, which is a fungal disease that can cause a white, powdery growth on leaves, stems, and flowers, and can eventually cause the plant to die. Meanwhile, pepper plants can be affected by bacterial wilt, which is a bacterial disease that can cause wilting, yellowing, and death of the plant. Cloves can be affected by a fungal disease called anthracnose, which can cause leaf spots, stem cankers, and fruit rot [3]. Proper identification and management of these diseases are crucial to ensure healthy plant growth and maximum yield. This can involve measures such as regular scouting for signs and symptoms of disease, maintaining proper sanitation practices, and implementing disease-resistant varieties of plants. Spices plants are an essential part of the culinary world, providing flavors and aromas that can make or break a dish. However, they are not immune to diseases that can harm their growth, yield, and quality. Identifying these diseases is crucial to managing them effectively and preventing their spread. Farmers and gardeners can use various methods to identify plant diseases, including books, the internet, and local garden 1 centers [4]. They can also take a small cutting of an infected plant or a photograph of it to their local garden center or extension service for diagnosis. It is important to note that accurate diagnosis requires careful observation of the plant's appearance, including the leaves, stems, flowers, and fruit [5]. One of the most common diseases in spices plants, as well as in other plants, is the mosaic virus [6]. It is a viral disease that affects more than 150 types of plants, including fruits, vegetables, and flowers. The virus causes mottled patterns of light and dark green on the leaves, which can distort the plant's growth and yield. Other diseases that can affect spices plants include bacterial and fungal infections. Symptoms of bacterial infections include wilting, yellowing, and brown spots on the leaves and stems [5]. Fungal infections can cause various symptoms, including powdery mildew, black spots, and rust-like growths on the leaves, stems, and flowers [6]. To identify diseases in spices plants, it is essential to look for signs and symptoms of the pathogen involved. Signs of plant diseases are physical evidence of the pathogen, such as fungal fruiting bodies, bacterial ooze, or viral particles [2]. Symptoms, on the other hand, are the visible effects of the disease on the plant, including a detectable change in color, shape, or function of the plant as it responds to the pathogen [5]. Identifying diseases in spices plants is crucial to managing them effectively and preventing their spread. Farmers and gardeners can use various methods, such as books, the internet, and local garden centers, to learn about plant diseases and how to identify them. Accurate diagnosis requires careful observation of the plant's appearance, including the leaves, stems, flowers, and fruit. Signs of plant diseases are physical evidence of the pathogen, while symptoms are the visible effects of the disease on the plant. By understanding these concepts and being observant, it is possible to prevent and control diseases in spices plants. Spices plants such as cinnamon, pepper, and cloves can be affected by various diseases that can cause visible symptoms on their leaves, branches, trunk, or roots. Understanding these visual symptoms is important for identifying and treating plant diseases. Here are some ways to understand the visual symptoms of spice plant diseases. To begin with, it is important to carefully observe the plant to identify any 2 physical evidence of the pathogen involved. Farmers and gardeners often use books and the internet to identify plant diseases. They can also take a small cutting of an infected plant (or a photograph of it) to a local garden center or extension service for help with diagnosis [7]. One of the most common visual symptoms of plant diseases is the formation of galls, which are growths or swellings found on leaves, stems, and/or roots. Galls can be caused by various factors such as bacteria, fungi, or insects. Scale insects can infest a wide variety of trees and shrubs, forming "bumps" on leaves and/or stems that can be flicked off with a knife point. Leaves or other surfaces can become shiny and sticky from the sugary excrement from the insects. Another visual symptom of plant diseases is a change in color, shape, or function of the plant as it responds to the pathogen. For example, a common viral disease that affects pepper plants is mosaic virus, which causes a mottled pattern of light and dark green on the leaves. This disease is mainly spread by insects, specifically aphids and leafhoppers [9]. In contrast, bacterial canker is a disease that affects fruit trees such as cherry, peach, and plum. The thick, liquid exudate is primarily composed of bacteria and is a sign of the disease, although the canker itself is composed of plant tissue and is a symptom [8]. Moreover, there are several types of visual symptoms that can indicate problems with the vascular tissue of the plant, which is responsible for transporting water, nutrients, and photosynthates throughout the plant. Wilting, yellowing, or browning of leaves can be a sign of vascular wilt diseases caused by fungi or bacteria. Root rot is another disease that affects the roots of plants, causing them to turn brown and mushy. These diseases can also cause stunting of the plant, reduced vigor, and premature death [8]. The presence of fungal fruiting bodies on leaves, stems, or branches is a visual symptom of fungal diseases such as powdery mildew or black spot. Powdery mildew is a common fungal disease that affects many different plants, including cinnamon, pepper, and cloves. It appears as a white or gray powder on the leaves and stems, causing them to become distorted and wilted [9]. 3 Technology has played an increasingly important role in the study and management of plant diseases. From the detection and diagnosis of plant diseases to the development of new management strategies, technology has greatly improved our understanding of plant diseases and how to combat them. One way in which technology is involved in plant disease is through the development of diagnostic tools [10]. Diagnostic technologies such as DNA sequencing, polymerase chain reaction (PCR), and enzymelinked immunosorbent assay (ELISA) have allowed for more accurate and rapid detection of plant pathogens. These tools can help identify the pathogen responsible for a plant disease and allow for the early detection of new or emerging diseases. Another way technology is involved in plant disease is using remote sensing and imaging [11]. Remote sensing involves the use of drones, satellites, and other technologies to collect data on plant health and disease. This data can be used to identify areas of stress in crops and monitor the spread of disease over large areas. Imaging technologies such as hyperspectral imaging can also be used to detect subtle changes in plant physiology that may indicate the presence of disease. Technology has also facilitated the development of new management strategies for plant diseases [12]. For example, precision agriculture techniques such as variable rate application of fertilizers and pesticides can reduce the risk of disease outbreaks by providing targeted application of inputs. Biological control agents, such as beneficial insects and microbes, can also be used to manage plant diseases. Furthermore, artificial intelligence (AI) and machine learning algorithms can help predict disease outbreaks and inform management decisions. 4 1.1 BACKGROUND & LITERATURE REVIEW 5 1.2 RESEARCH GAP Features Systems Proposed System 6 1.3 RESEARCH PROBLEM 7 1.4 RESEARCH OBJECTIVES 1.4.1 MAIN OBJECTIVE The main objective of this mobile-based solution is to provide a tool for identifying diseases of selected spice plants, including cinnamon, pepper, and cloves. By inputting an image of the plant, leaf, root, branch, or trunk to the application, the tool will identify the disease and show possible remedies to control it. 1.4.2 SPECIFIC OBJECTIVES To study and understand visual symptoms of leaves, plant, root, branches, or trunk of spice plants: This objective involves studying the visual symptoms of the selected spice plants, which can include changes in color, shape, texture, or structure that indicate the presence of a disease. By studying these visual symptoms, the researchers can develop a comprehensive understanding of the diseases that commonly affect the selected spice plants and identify the most effective methods for diagnosing and treating these diseases. To study and understand how to capture images on spice plants to train model: This objective involves studying the different techniques and methods for capturing clear and accurate images of the spice plants. High-quality images are essential for training the model effectively, and researchers may need to use specialized equipment, such as cameras with high resolution and specialized lenses or lighting, to capture these images. The researchers may also need to develop specific protocols for capturing images of different parts of the plant, such as leaves, branches, or roots. To study and understand types of techniques which are related to image processing, deep learning, and machine learning to train a model: This objective involves studying various techniques related to image processing, deep learning, and machine learning to train the model to accurately identify the disease. Image processing techniques can be used to enhance the quality of the input images, 8 while deep learning and machine learning techniques can be used to analyze the images and identify patterns or features that indicate the presence of a disease. The researchers may need to experiment with different techniques and approaches to determine the most effective methods for training the model. To study and understand how to implement a mobile application to identify diseases and show possible remedies to control the diseases: This objective involves studying how to develop and implement a mobile application that can identify the disease and provide possible remedies to control it. The application may need to be user-friendly and accessible, with clear instructions and intuitive interfaces. The researchers may also need to identify reliable sources of information on possible remedies for the identified diseases and integrate this information into the application. Additionally, the application may need to incorporate feedback mechanisms to allow users to report their experiences and provide additional information on the effectiveness of different remedies. 9 2 2.1 METHODOLOGY System Architecture Figure 2.1 System overview diagram The main objective of the project is to provide farmers with a smart mobile solution that can help them diagnose and manage the diseases of certain spice plants. This is achieved by using the suggested method, which allows farmers to input a photograph of the surface of the spice plant into the mobile application. The application then uses image processing and deep learning techniques to identify the disease present in the spice plant and recommend effective methods of disease control. The proposed system is designed to minimize the severe impact of plant diseases by visualizing the spread of illnesses according to their severity and notifying farmers whenever an infected tree is discovered. This way, farmers can take prompt action to prevent the spread of the disease and minimize the damage to their crops. Overall, the project aims to provide a simple and user-friendly solution for farmers to identify and manage the diseases of their spice plants, ultimately helping them to improve their crop yields and profitability. 10 2.2 2.2.1 Development Process Component Figure 2.2 Component Diagram This component of the proposed system is responsible for processing the captured image of the spice plant and identifying any diseases or symptoms present in the leaves. It uses a Convolutional Neural Network (CNN) to accurately identify the edges and shapes on the leaf images. The CNN is trained to identify various disease patterns and classify them accordingly. If the leaf turns brown, the CNN uses color segmentation to identify brown color patterns in the disease, and extract features to detect the symptoms. This allows for more accurate 11 identification of the disease compared to other basic image processing and machine learning techniques. To further improve the image processing, the OpenCV library is used to denoise the images before processing. OpenCV is a widely-used computer vision library that provides various functions for image processing, including fastNlMeansDenoising and fastNlMeansDenoisingColored, which remove noise from images and improve their clarity. The proposed system specifically uses the MobileNetV2 architecture from scratch to solve the problem of identifying diseases in spice plants. This architecture is optimized for mobile devices and has been trained on large datasets, making it a suitable choice for the proposed mobile application. Overall, this component plays a crucial role in accurately identifying diseases in spice plants and providing effective solutions to farmers. 12 2.2.2 Tools and technologies React Native React Native is a cross-platform mobile application development framework that can be used to develop the proposed system for identifying diseases and showing possible remedies to control the diseases of selected spice plants. React Native allows developers to build mobile applications that can run on both iOS and Android platforms with a single codebase, which makes it a cost-effective solution for mobile application development. To achieve the main objective, React Native can be used to create a mobile application that allows users to input an image of the plant, leaf, root, branch, or trunk of selected spice plants, such as cinnamon, pepper, and cloves, to identify the disease. The application can be designed to process the input image using image processing, deep learning, and machine learning techniques to accurately identify the disease and suggest possible remedies for disease control. React Native can also be used to implement specific objectives, such as studying and understanding the visual symptoms of leaves, plant, root, branches, or trunk of spice plants, and how to capture images on spice plants to train the model. The application can provide features that allow users to learn about the visual symptoms of diseases in selected spice plants and how to capture clear and accurate images of the plants for training the model. Additionally, React Native can be used to implement the study and understanding of types of techniques related to image processing, deep learning, and machine learning to train a model. The mobile application can use these techniques to train the model to identify diseases accurately, and the application can be designed to update the model regularly with new data to improve its accuracy. 13 Image Processing and Convolutional Neural Networks (CNN) play a crucial role in the identification and management of diseases in selected spice plants like cinnamon, pepper, and cloves. The process involves capturing an image of the plant or its parts (leaf, root, branch, or trunk) and uploading it to the mobile application for analysis. The image is then processed using Image Processing techniques, which are used to extract relevant features from the image that can help identify the disease. CNN is a deep learning technique that is commonly used in image analysis applications. It is used to train the model to identify patterns in the images that correspond to specific diseases. CNN is an accurate method for detecting edges, shapes, and patterns on leaf images. Once the model has been trained, it can be used to identify diseases in new images with a high degree of accuracy. The mobile application will use the trained model to identify the disease based on the features extracted from the image. The identified disease will then be matched with a database of possible remedies to suggest the most effective treatment. Overall, Image Processing and CNN are essential components of the mobile application for identifying and managing diseases in selected spice plants. They allow for quick and accurate identification of diseases, which can help prevent the spread of the disease and minimize the impact on crop yields. 14 Image Processing – CNN Image Processing and Convolutional Neural Networks (CNN) play a crucial role in the identification and management of diseases in selected spice plants like cinnamon, pepper, and cloves. The process involves capturing an image of the plant or its parts (leaf, root, branch, or trunk) and uploading it to the mobile application for analysis. The image is then processed using Image Processing techniques, which are used to extract relevant features from the image that can help identify the disease. CNN is a deep learning technique that is commonly used in image analysis applications. It is used to train the model to identify patterns in the images that correspond to specific diseases. CNN is an accurate method for detecting edges, shapes, and patterns on leaf images. Once the model has been trained, it can be used to identify diseases in new images with a high degree of accuracy. The mobile application will use the trained model to identify the disease based on the features extracted from the image. The identified disease will then be matched with a database of possible remedies to suggest the most effective treatment. Overall, Image Processing and CNN are essential components of the mobile application for identifying and managing diseases in selected spice plants. They allow for quick and accurate identification of diseases, which can help prevent the spread of the disease and minimize the impact on crop yields. 15 2.3 COMMERCIALIZATION ASPECTS OF THE PRODUCT Market Potential: The market potential for a disease identification application for spice plants is significant, as farmers and growers are always looking for new and innovative ways to manage and prevent diseases. This application could provide a valuable tool for growers to quickly identify and treat diseases, potentially leading to increased crop yields and profits. Licensing and Distribution: The application could be licensed and distributed to agricultural companies, nurseries, and farming organizations. These companies could then distribute the application to their clients or members, providing a value-added service and potentially increasing their revenue. Partnership Opportunities: The development of this application could lead to partnership opportunities with companies that specialize in agriculture, such as fertilizer or pesticide companies. These companies could use the application to identify specific diseases and provide targeted solutions to their customers. Subscription Model: The application could be offered as a subscription-based service, where growers pay a monthly or yearly fee to access the disease identification and treatment information. This model could provide a recurring revenue stream and potentially increase customer loyalty. Expansion into Other Crops: Once the application is developed and tested for spice plants, it could potentially be expanded to include other crops, such as fruits and vegetables. This would broaden the market potential and potentially increase the application's revenue. 16 2.4 2.4.1 Test case ID T_1 TESTING & IMPLEMENTATION Testing Test case Test input Values Name Test Procedure T_2 17 Expected Output Actual Result Test Resul t T_3 T_4 18 T_5 T_6 Table 2.1 TEST CASES 19 3 3.1 RESULTS & DISCUSSION RESULTS 20 3.2 RESEARCH FINDINGS 21 3.3 DISCUSSION 22 4 SUMMARY OF EACH STUDENT’S CONTRIBUTION Member Components Tasks 23 5 CONCLUSION 24 6 REFERENCES 25 7 APPENDICES 26