LEAF DISEASE DETECTION SYSTEM

Abstract

ABSTRACT “LEAF DISEASE DETECTION SYSTEM” The present invention relates to a leaf disease detection system [100] comprising an input module [102] with a camera and multiple image sensors, capturing images of tree leaves, a pre-processing unit [104] performs histogram equalization and contrast enhancement on the images to enhance quality, a segmentation module [106] then segments the pre-processed image into distinct regions for analysis using a BIRCH technique, an artificial intelligence (AI) module, connected to the input module [102], receives the images and uses pre-saved data to analyze them, the AI module [108] extracts features from segmented images, such as color, pixel features, center symmetric local ternary patterns (CSLTP), and center symmetric local derivative patterns (CSLDP) and the AI module [108] trains at least one AI model based on the analyzed images and extracted features, deploying the trained model to determine disease presence in the tree leaves. Refer to Figure 1

Specification

Description:"LEAF DISEASE DETECTION SYSTEM"
FIELD OF THE INVENTION
[0001] The present invention relates to the field of a leaf disease detection system that is capable of accurately detecting and identifying diseases in tree leaves to help in early diagnosis, improve crop health, and support agricultural productivity.
BACKGROUND OF THE INVENTION
[0002] In the field of agriculture, the health monitoring of crops is crucial for ensuring sustainable yields and minimizing losses. Detecting diseases in plants at an early stage may prevent widespread damage, reducing the need for excessive pesticide use and promoting better yield outcomes. Monitoring plant health is vital to maintaining food security and addressing the challenges faced by farmers globally, particularly in managing crop diseases that significantly impacts productivity and quality.
[0003] Traditionally, leaf disease detection is performed through manual inspection by experts or trained personnel, who visually assess plants for disease symptoms such as discoloration, spots, or wilting. This method, while straightforward, is highly dependent on the observer's expertise and is limited by human error, variability, and fatigue. Furthermore, manual inspection is time-consuming and impractical for large-scale operations, especially when diseases need to be detected in real time for timely intervention.
[0004] The traditional approach faces several challenges. It is not only labor-intensive but also lacks consistency and efficiency, often leading to delayed detection, misdiagnosis, and inappropriate treatments. This delay allows diseases to progress unchecked, reducing the effectiveness of control measures and impacting crop quality. Moreover, manual inspections are generally reactive rather than proactive, addressing symptoms only once they are visible, rather than detecting underlying disease markers early. These limitations underscore the need for a more advanced, accurate, and efficient solution for leaf disease detection in modern agriculture.

SUMMARY OF THE INVENTION
[0005] In view of the foregoing disadvantages inherent in the prior art, the general purpose of the present disclosure is to provide leaf disease detection system, to include all advantages of the prior art, and to overcome the drawbacks inherent in the prior art.
[0006] Some of the objects of the present disclosure, which at least one embodiment herein satisfies, are as follows:
[0007] An object of the present disclosure is to ameliorate one or more problems of the prior art or to at least provide a useful alternative. An object of the present disclosure is to provide a leaf disease detection system.
[0008] Another object of the present disclosure is to provide a leaf disease detection system that is capable of providing an advanced solution for detecting diseases in plant leaves, enhancing early identification of potential health issues in crops.
[0009] Another object of the present disclosure is to provide a leaf disease detection system that is capable of improving the accuracy of plant health assessment by utilizing systematic analysis methods that go beyond traditional manual observation.
[0010] Another object of the present disclosure is to provide a leaf disease detection system that is capable of offering an automated, scalable approach suitable for large-scale agricultural operations, minimizing the need for human intervention.
[0011] Another object of the present disclosure is to provide a leaf disease detection system that is capable of supporting real-time analysis for prompt feedback on plant health, helping farmers make informed decisions on disease management and crop treatment.
[0012] Yet another object of the present disclosure is to provide a leaf disease detection system that is capable of generating comprehensive health reports, including insights into disease location and severity, to better guide crop maintenance practices.
[0013] Other objects and advantages of the present disclosure will be more apparent from the following description, which is not intended to limit the scope of the present disclosure.
[0014] In view of the above objects, in one aspect, the current disclosure provides a leaf disease detection system that is a robust and novel phishing-resistant authentication tool.
[0015] The leaf disease detection system of the present disclosure facilitates an input module with a high-resolution camera and multiple sensors to capture both visible and infrared images of leaves. Then a pre-processing unit enhances the images by applying histogram equalization and adaptive contrast techniques for clearer visuals. Following this, a segmentation unit divides the images based on texture and color, utilizing the BIRCH clustering technique for effective region separation. Further, an AI module analyzes the segmented images, extracting features like color, pixel details, and patterns (such as CSLTP and CSLDP) for disease identification. The AI module then trains an AI model with the extracted features and applies the trained model to detect diseases in real-time. The AI model classifies the specific disease type, provides immediate feedback, and generates a comprehensive report, covering the location and severity of the disease. Furthermore, in order to optimize performance, the AI module includes evaluation mechanisms using both parametric and non-parametric analyses, as well as a self-upgraded honey badger optimization (SUHBO) model, enhancing detection accuracy and system responsiveness.



BRIEF DESCRIPTION OF DRAWING
[0016] The foregoing summary, as well as the following detailed description of various embodiments, is better understood when read in conjunction with the drawings provided herein. For the purposes of illustration, there are shown in the drawings exemplary embodiments; however, the presently disclosed subject matter is not limited to the specific methods and instrumentalities disclosed.
[0017] Figure 1 illustrates an exemplary block diagram of a leaf disease detection system, in accordance with exemplary implementations of the present disclosure;
[0018] Figure 2 illustrates an exemplary method flow diagram depicting method of operation of the leaf disease detection system, in accordance with exemplary implementations of the present disclosure; and
[0019] Figure 3 illustrates an exemplary flow diagram of a multiple stages of a leaf disease detection system, in accordance with exemplary implementations of the present disclosure.
[0020] Like reference numerals refer to like parts throughout the description of several views of the drawing.
DETAILED DESCRIPTION OF THE INVENTION
[0021] Embodiments are provided so as to thoroughly and fully convey the scope of the present disclosure to the person skilled in the art. Numerous details are set forth, relating to specific components, and methods, to provide a complete understanding of embodiments of the present disclosure. It will be apparent to the person skilled in the art that the details provided in the embodiments should not be construed to limit the scope of the present disclosure. In some embodiments, well-known processes, well- known apparatus structures, and well-known techniques are not described in detail.
[0022] The terminology used, in the present disclosure, is only for the purpose of explaining a particular embodiment and such terminology shall not be considered to limit the scope of the present disclosure. As used in the present disclosure, the forms "a," "an," and "the" may be intended to include the plural forms as well, unless the context clearly suggests otherwise. The terms "comprises," "comprising," "including," and "having," are open-ended transitional phrases and therefore specify the presence of stated features, integers, steps, operations, elements, modules, units and/or components, but do not forbid the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. The particular order of steps disclosed in the method and process of the present disclosure is not to be construed as necessarily requiring their performance as described or illustrated. It is also to be understood that additional or alternative steps may be employed.
[0023] The following detailed description should be read with reference to the drawings, in which similar elements in different drawings are identified with the same reference numbers. The drawings, which are not necessarily to scale, depict illustrative embodiments and are not intended to limit the scope of the disclosure.
[0024] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed. In this application, the use of the singular includes the plural, the word "a" or "an" means "at least one", and the use of "or" means "and/or", unless specifically stated otherwise. Furthermore, the use of the term "including", as well as other forms, such as "includes" and "included", is not limiting. Also, terms such as "element" or "component" encompass both elements and components comprising one unit and elements or components that comprise more than one unit unless specifically stated otherwise.
[0025] Furthermore, the term "module", as used herein, refers to logic embodied in hardware or firmware, or to a collection of software instructions, written in a programming language, such as, Java, C, C++, python, or assembly. One or more software instructions in the modules can be embedded in firmware, such as in an EPROM. The modules described herein can be implemented as either software and/or hardware modules and can be stored in any type of non-transitory computer-readable medium or other storage device. Some non-limiting examples of non-transitory computer-readable media include CDs, DVDs, BLU-RAY, flash memory, and hard disk drives.
[0026] Referring to Figure 1, an exemplary block diagram of a leaf disease detection system [100], is shown, in accordance with exemplary implementations of the present disclosure. The leaf disease detection system [100] comprises at least one input module [102], at least one pre-processing unit [104], at least one segmentation module [106], and at least one artificial intelligence (AI) module [108]. Also, all of the components of the leaf disease detection system [100] are assumed to be connected to each other unless otherwise indicated below. As shown in the figures all components shown within the leaf disease detection system [100] should also be assumed to be connected to each other. Also, in Figure 1, only a few units are shown, however, the leaf disease detection system [100] may comprise multiple such units or the leaf disease detection system [100] may comprise any such numbers of said components, as required to implement the features of the present disclosure.
[0027] The leaf disease detection system [100] comprises the input module [102] that utilizes a high-resolution digital camera and multiple image sensors. The mentioned components capture detailed images of tree leaves across both visible and infrared spectrums. Infrared imaging enables the detection of variations in leaf health that may not be visible in standard light, offering a broader scope of analysis and improved disease detection capabilities. The captured data is critical, forming the foundation of the detection process by ensuring that images include all relevant details necessary for analysis.
[0028] The captured images are sent to the pre-processing unit [104] where they undergo initial modifications to enhance quality. The pre-processing unit [104] uses techniques such as histogram equalization to balance the distribution of pixel intensities and contrast enhancement to bring out subtle details. Specifically, local adaptive histogram equalization is applied, which adjusts contrast dynamically based on localized image regions, which allows for greater precision in capturing variations that might indicate disease.
[0029] Following pre-processing, the images are segmented using the segmentation module [106]. The segmentation module [106] uses the Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) algorithm, which collects the image into sections based on visual characteristics like color and texture. The BIRCH algorithm creates Clustering Feature (CF) trees that aggregate image areas with similar properties while retaining granular details necessary for subsequent steps. By organizing image sections into a hierarchical tree with sub-clusters, the segmentation module [106] process allows the leaf disease detection system [100] to analyze specific image regions accurately. This segmentation module [106] divides the image in a structured way, enabling more precise disease diagnosis by focusing on significant patterns and regions.
[0030] Once segmentation by the segmentation module [106] is complete, the leaf disease detection system [100] moves into the feature extraction phase. Here, various unique features are extracted from the segmented images, including color features, pixel-based features, Center Symmetric Local Ternary Patterns (CSLTP), and Center Symmetric Local Derivative Patterns (CSLDP). The CSLTP feature extraction method examines relationships between a central pixel and its diagonally paired surrounding pixels, which helps in identifying leaf patterns indicative of disease. CSLDP, an advancement on CSLTP, captures even finer details by analyzing directional changes in pixel intensities, thereby enhancing the detection process.
[0031] The processed and segmented data is then analyzed by the AI module [108]. The AI module [108] is built around a hybrid classifier combining Deep-Maxout and Bidirectional Long Short-Term Memory (Bi-LSTM) models. Deep-Maxout neurons activate based on maximum values across multiple candidates, effectively distinguishing between complex patterns. Bi-LSTM adds temporal depth by analyzing data in both forward and backward directions, taking into account both preceding and succeeding information, which improves the leaf disease detection system [100]'s ability to recognize patterns in time-series data. Additionally, the model is optimized through Self-Upgraded Honey Badger Optimization (SUHBO), an innovative technique that dynamically tunes model parameters to enhance detection accuracy. This optimization minimizes errors during training, leading to more reliable disease predictions and efficient model performance.
[0032] The AI module [108] uses the features extracted from images to classify the presence and type of disease in the leaves. By combining Bi-LSTM's sequential processing abilities with Deep-Maxout's nuanced activation method, the leaf disease detection system [100] achieves a robust classification mechanism. After the classification, the leaf disease detection system [100] displays the results on a user interface connected to a computing device, allowing users to view the diagnostic information promptly. The AI module [108] also generates a detailed report summarizing the tree's health status, including the severity of disease, precise location of affected regions, and any notable disease characteristics.
[0033] In real-time operation, the AI module [108] enables immediate feedback on the health status of tree leaves. As a result, users can make timely decisions on necessary interventions. Additionally, the AI module [108] incorporates a performance evaluation framework that employs both parametric and non-parametric analysis, assessing the model's accuracy and efficiency, which helps in refining the leaf disease detection system [100] over time.
[0034] Referring to Figure 2, an exemplary method flow diagram [200] for operating the operation of the leaf disease detection system [100], in accordance with exemplary implementations of the present disclosure is shown. In an implementation the method [200] is performed by the operation of the leaf disease detection system [100].
[0035] Also, as shown in Figure 2, the method [200] initially starts at step [202].
[0036] At step [204], the method comprises capturing, via an input module [102] having a camera and a plurality of image sensors, one or more images of one or more leaves of a tree. The mentioned components capture detailed images of tree leaves across both visible and infrared spectrums. Infrared imaging enables the detection of variations in leaf health that may not be visible in standard light, offering a broader scope of analysis and improved disease detection capabilities.
[0037] At step [206], the method comprises performing, via a pre-processing unit [104], histogram equalization and contrast enhancement on the captured images to improve image quality. The pre-processing unit [104] uses techniques such as histogram equalization to balance the distribution of pixel intensities and contrast enhancement to bring out subtle details.
[0038] At step [208], the method comprises segmenting, via a segmentation module [106], the pre-processed image into distinct regions for analysis using a BIRCH technique, which clusters the image into sections based on visual characteristics like color and texture. The BIRCH algorithm creates Clustering Feature (CF) trees that aggregate image areas with similar properties while retaining granular details necessary for subsequent steps. By organizing image sections into a hierarchical tree with sub-clusters, the segmentation process allows the leaf disease detection system [100] to analyze specific image regions accurately.
[0039] At step [210], the method comprises receiving, via an artificial intelligence (AI) module [108] communicatively coupled with the input module [102], the one or more images.
[0040] At step [212], the method comprises analysing, via the AI module [108], the one or more images using a pre-saved data.
[0041] At step [214], the method comprises extracting, via the AI module [108], features from the segmented image, including color features, pixel features, center symmetric local ternary patterns (CSLTP), and center symmetric local derivative patterns (CSLDP). The CSLTP feature extraction method examines relationships between a central pixel and its diagonally paired surrounding pixels, which helps in identifying leaf patterns indicative of disease. CSLDP, an advancement on CSLTP, captures even finer details by analyzing directional changes in pixel intensities, thereby enhancing the detection process.
[0042] At step [216], the method comprises training, via the AI module [108], at least one AI model based on the analysed one or more images and extracted features. The AI module [108] is built around a hybrid classifier combining Deep-Maxout and Bidirectional Long Short-Term Memory (Bi-LSTM) models. Deep-Maxout neurons activate based on maximum values across multiple candidates, effectively distinguishing between complex patterns. Bi-LSTM adds temporal depth by analyzing data in both forward and backward directions, taking into account both preceding and succeeding information, which improves the leaf disease detection system [100]'s ability to recognize patterns in time-series data.
[0043] At step [218], the method comprises deploying, via the AI module [108], the trained AI model to determine disease in the one or more leaves of the tree.
[0044] The method herein ends at step [220].
[0045] Referring to Figure 3, an exemplary flow diagram [300] of multiple stages of a leaf disease detection system [100], in accordance with exemplary implementations of the present disclosure is shown.
[0046] At stage [302], the leaf disease detection system [100] is in an input stage where high-resolution images are captured by a digital camera and multiple sensors, covering both visible and infrared spectrums to enhance detection of leaf health variations.
[0047] At stage [304], the leaf disease detection system [100] is in pre-processing stage, where the captured images undergo histogram equalization and contrast enhancement to improve clarity, ensuring that subtle details are accentuated.
[0048] At stage [306], the leaf disease detection system [100] is in segmentation stage, where the BIRCH algorithm clusters the image based on visual traits like color and texture, organizing it into structured segments for focused analysis.
[0049] At stage [308], the leaf disease detection system [100] is in feature extraction stage, where the identification of specific characteristics, such as color, pixel distribution, and more complex patterns like CSLDP and CSLTP, which detect unique leaf traits associated with disease is done.
[0050] At stage [310], the leaf disease detection system [100] is in hybrid classification stage, where a Deep Max-Out and Bi-LSTM model, with weights optimized using the SUHBO algorithm is utilized to accurately diagnoses disease types based on extracted features, providing an output for disease detection and severity assessment.
[0051] While considerable emphasis has been placed herein on the specific features of the preferred embodiment, it will be appreciated that many additional features can be added and that many changes can be made in the preferred embodiment without departing from the principles of the disclosure. These and other changes in the preferred embodiment of the disclosure will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the disclosure and not as a limitation.
[0052] While the invention has been described in connection with what is presently considered to be the most practical and various embodiments, it is to be understood that the invention is not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements.
[0053] The embodiments described above are intended only to illustrate and teach one or more ways of practicing or implementing the present invention, not to restrict its breadth or scope. The actual scope of the invention, which embraces all ways of practicing or implementing the teachings of the invention, is defined only by the following claims and their equivalents.
, Claims:I/WE CLAIM:
1. A leaf disease detection system [100], comprising:
an input module [102] having a camera and a plurality of image sensors, configured to capture one or more images of one or more leaves of a tree;
a pre-processing unit [104] configured to perform histogram equalization and contrast enhancement on the captured images to improve image quality;
a segmentation module [106] configured to segment the pre-processed image into distinct regions for analysis using a BIRCH technique;
an artificial intelligence (AI) module [108] communicatively coupled with the input module [102], wherein the AI module [108] is configured to:
receive the one or more images,
analyse the one or more images using a pre-saved data,
extract features from the segmented image, including color features, pixel features, center symmetric local ternary patterns (CSLTP), and center symmetric local derivative patterns (CSLDP),
train at least one AI model based on the analysed one or more images and extracted features,
deploy the trained AI model to determine disease in the one or more leaves of the tree.
2. The leaf disease detection system [100] as claimed in claim 1, wherein the camera is a high-resolution digital camera capable of capturing both visible and infrared images of the tree leaves.

3. The leaf disease detection system [100] of claim 1, wherein the pre-processing unit [104] further enhances the contrast by applying a local adaptive histogram equalization technique.

4. The leaf disease detection system [100] of claim 1, wherein the segmentation module [106] is configured to divide the image into regions based on texture and color differences using the BIRCH technique.

5. The leaf disease detection system [100] of claim 1, wherein the AI module [108] is further configured to classify the type of disease based on the trained model and display the result on a user interface installed within a computing unit.

6. The leaf disease detection system [100] of claim 1, wherein the AI module [108] is configured to generate a report on the health status of the tree, including the location and severity of detected diseases.

7. The leaf disease detection system [100] of claim 1, wherein the AI module [108] is capable of operating in real-time, providing immediate feedback on leaf health and disease detection

8. The leaf disease detection system [100] of claim 1, wherein the AI module [108] is configured to evaluate the performance of the AI model using both parametric and non-parametric analysis.

9. A method for operating the leaf disease detection system [100] as claimed in claim 1, wherein the method comprising:

capturing, via an input module [102] having a camera and a plurality of image sensors, one or more images of one or more leaves of a tree;
performing, via a pre-processing unit [104], histogram equalization and contrast enhancement on the captured images to improve image quality;
segmenting, via a segmentation module [106], the pre-processed image into distinct regions for analysis using a BIRCH technique;
receiving, via an artificial intelligence (AI) module [108] communicatively coupled with the input module [102], the one or more images;
analysing, via the AI module [108], the one or more images using a pre-saved data;
extracting, via the AI module [108], features from the segmented image, including color features, pixel features, center symmetric local ternary patterns (CSLTP), and center symmetric local derivative patterns (CSLDP);
training, via the AI module [108], at least one AI model based on the analysed one or more images and extracted features; and
deploying, via the AI module [108], the trained AI model to determine disease in the one or more leaves of the tree.
10. The method for operating the leaf disease detection system [100] as claimed in claim 9, wherein the AI module [108] utilizes a self upgraded honey badger optimization (SUHBO) model to improve detection accuracy.

Documents