LIST OF APPENDICES
2 Source code
4 Respondent photo Documentation
5 User manual
6 Curriculum vitae (Researcher)
LIST OF TABLES
1 Hardware Requirements
2 Software Requirements
3 Budgetary Requirements
4 Respondents of the Study
5 Likert Scale of Functionality
6 Likert Scale of Reliability
7 Testing of the system day 3
8 Testing of the system day 4
9 Testing of the system day 5
10 Confusion matrix
11 Descriptive rating of the Respondents of Remote-control camera
12 Descriptive ratings of Camera that slides along the track
13 Descriptive rating of the controlled camera system
14 Descriptive rating of the healthy Solid Trichoderma Ready to harvest
15 Descriptive rating of Classify healthy and contaminated
16 Descriptive of the system would not create an error
17 Descriptive rating of the system resumes working and restores lost data after failure
18 Descriptive rating of the system did not rely to any other hardware
19 Descriptive ratings of the system would not cause any error and malfunction
LIST OF FIGURES
1 Conceptual Framework of the study
2 Locale map of the study of Davao del sur state college
3 Waterfall Diagram
4 Theoretical Framework of the study
5 The operational Framework of Trichoderma based on color code Texture
6 Flow Chart
7 Remote control to run the motorized camera dolly
8 Camera that slides along the track, ensuring smooth and steady movement
9 Captured image using controlled camera
10 Determine the healthy solid Trichoderma culture ready to harvest
11 Flow chart
12 Pseudocode of healthy solid Trichoderma ready to harvest
13 Classify difference of solid Trichoderma between healthy and contaminated through color recognition
14 Flow chart
15 Pseudocode of Classification between healthy and contaminated
16 Testing manual efficiency
17 Generate software
18 Training loss
19 Training accuracy
20 Model screen
21 Prototype of the desktop application system
Background of the Study
A color code, also known as a color scheme, is a method of showing details using various colors. The parameters ambient, diffuse, specular, opacity, and color describe a solid color texture. These color-coded textures are used to determine the different colors of Trichoderma. Trichoderma is a free-living fungus that is highly interactive in the root, soil, and foliar habitats and is offered to organic producers as a wide range of Trichoderma-based products Plant diseases can be controlled with biopesticides, which are biological insecticides. Trichoderma species are also important because they stimulate plant development while inhibiting the growth of plant diseases.(Palm, 2016)
Nowadays, Trichoderma species Due to its well-known biological control mechanism, it's been frequently employed in agricultural applications. Due to the usage of this microbial inoculant in Trichoderma-based products, researchers are interested in learning more about the many prospective advantages of Trichoderma. Plant diseases, growth, breakdown, and bioremediation are all discussed. This article will also look at how they produce secondary metabolites in the agroecosystem. According to (Zin and Badaluddin, 2020), these groundbreaking discoveries bring significant benefits to the agriculture industry in terms of environmentally friendly farming practices.
However, the Trichoderma laboratory in charge at Davao del Sur State College uses only their naked eyes to check the difference between good Trichoderma and contaminated based on the color of the solid Trichoderma. The institution’s Trichoderma laboratory employs one person in-charge and one trained agriculturalist as the head of the laboratory. The procedure is to acquire a small amount of solid Trichoderma, also known as the mother culture in the solid Trichoderma. And move it into the solid Trichoderma container Petri plate (PDA). Pre-heat the transfer needle in the alcohol lamp within one minute to transfer the small amount of solid Trichoderma before using it to keep the Solid Trichoderma from being contaminated. After that process, it must put in the petri plate into the laboratory table/shelve or containing area and ensure that the petri plate is not shaken. The growth of Trichoderma inside the petri plate will not be destructed, and not to touch it. After that, set it aside for a week. The in-charge of the laboratory could then harvest the good solid Trichoderma. In classifying harvesting in solid Trichoderma, it adheres to the product's Quality Control level and automate the manual classification process in time for harvest. Using the sense of sight identify solid Trichoderma in each container as a manual operation, will have a lower level of accuracy. It would be difficult to tell whether the solid Trichoderma is safe or not. As a result, the scientist is working on an application that will enable the person to quickly evaluate and identify the exact state of solid Trichoderma in the lab.(3 and Gallen, 2017)
Therefore, the proponent develop the Trichoderma Classification System based on Color Code Texture of Potato Dextrose Agar Solid (PDA). Using TensorFlow, it has a prototype real track to capture an image and system classify to determine the solid Trichoderma. Also, it automates the manual process by taking a picture of solid Trichoderma with the system's camera. On the other hand, the solid Trichoderma classification process is more reliable when it comes to harvesting, which will help to maintain the product's quality control. The camera will shift automatically using a motorized camera dolly. The captured images automatically recognized and display the result on the computer monitor.
Objective of the Study
The overall goal of this research is to create a “Trichoderma Classification System Based on Color Code Texture of Potato Dextrose Agar Solid (PDA) Using TensorFlow.
Specifically, on the following:
1. To develop a Prototype with the following functionality.
1.1 Remote control to run the motorized camera dolly.
1.2 Camerathat slides along the track, ensuring smooth and steady movement.
1.3 Capture images using controlled camera system.
2. To develop a System has the following functionality.
2.1 Determine the healthy Solid Trichoderma culture Ready to harvest.
2.2 Classify difference of solid Trichoderma between good or contaminated through color recognition.
3. To Evaluate Trichoderma Classification System Based on Color Code Texture of Potato Dextrose Agar Solid (PDA) Using TensorFlow with Prototype in terms of:
Significance of the Study
The study of Trichoderma classification system based on color code texture of Potato Dextrose Agar Solid (PDA) using TensorFlow with a prototype will be beneficial to the following future researcher, quality control, and the laboratory in charge of the Trichoderma laboratory of Davao del Sur State College.
Future Researcher. This project would provide baseline data and a rich source of a related study that they will be conducting in the future, students and teachers alike in terms of Information Technology innovations. Relevant information from this study would serve as a guide for future developers who may get involved in the same project or related program.
Quality Control. This study will help maintain the quality of Trichoderma in the laboratory, and will help not to enter microorganisms in the laboratory and infect others.
Laboratory In-charge. This study will be the basis for Establishing quality standards, designing operations, quality, and troubleshooting processes, assuring staff compliance, certifying instrument performance, scheduling equipment and replacement, servicing, and repair are all ways to improve laboratory equipment performance.
Scope and Limitations of the Study
The scope of Trichoderma classification system based on color code texture of Potato Dextrose Agar Solid (PDA) using Tensor Flow with prototype is accessible only by the desktop/computer. The use of this system is to classify and determine the Trichoderma by the time of harvesting. Only the color is used to organize and define the excellent Trichoderma like white, green, and dark-green. If the system classifies the different color, it will automatically identify as contaminated Trichoderma. The system will be used after five days or six days of planting the solid Trichoderma. The solid Trichoderma is placed in Petri plates by the researcher. A laboratory is used to perform or analyze the research. The researcher is able to catch 2000-3000 images. This project's data was comprised of captured photos. The laboratory in charge, the agriculturist, and the quality control officer are chosen as experts for this project to determine the reliability and functionality of the application. The researcher presents a standardized survey only for the project. This research is being studied to ensure the consistency of solid Trichoderma.
Definition of Terms
In the context of this study, the following terms were defined either theoretically and/or operationally:
Color code Texture. The colors red, green, and blue are represented by hexadecimal triplets, such as '255' red, '0' green, and '0' blue in the color red. And there's no image mapping or procedural texture mapping, so texture definition is just a solid color.
Trichoderma. This one free-living fungus that is highly interacting in root, soil, and foliar settings (Singh et al., 2014) and is accessible to organic producers as a wide range of Trichoderma-based biopesticides for plant disease control.
TensorFlow. Anyone, from start to end, may utilize this open-source machine learning platform. This is where you can find TensorFlow's diverse ecosystem of tools, libraries, and community resources. TensorFlow provides Machine Learning APIs. TensorFlow is also faster than rival Deep Learning frameworks when it comes to compilation.
Program. This refers to a set of operations that is made to function for a specific task.
Prototype. It's utilized to put a fresh design to the test so that analysts and system users can be more accurate. It is the stage that occurs between the formalization and evaluation of a notion.
Functionality. The number of processes that a computer or other electronic system can perform.
Reliability. One of the most important aspects of a successful psychological evaluation. After all, it wouldn't be very useful if a test was inconsistent and produced different findings every time.
Review of Related Study
Color sensing and image processing-based automatic soybean plant foliar disease severity detection and estimation
The study effort presented in this publication, according to (Shrivastava et al., 2015), focuses on cultivation difficulties and highlights the influence of various soya plant foliar diseases on yield. It has been built a completely automated illness diagnosis and level estimate system based on color image sensing and processing. Farmers' awareness may be increased in various agricultural areas, such as watershed management, natural disaster management, and insect and disease control, using data engineering and associated tools and methodologies. Farmers' awareness may be increased in various agricultural areas, such as watershed management, natural disaster management, and insect and disease control, using data engineering and associated tools and methodologies. The study presented in this paper focuses on the advancement of color sensors and image processing techniques for automated delivery of soy bugs and illnesses. Commercial farmers in industrialized nations have employed the right knowledge base and the newest technology to deal with this tough scenario because of high input costs, however this is not the case in developing countries like India. The objective of this research is to bridge the gap between commercial farmers in underdeveloped nations and supply cutting-edge technology.
An improved Canny edge detection algorithm for color image
This traditional approach, according to (Geng et al., 2012), is unable to handle color pictures, and adaptive parameter selection is challenging. In the processing of gray pictures, the traditional approach of Canny edge detection is often utilized. This study proposes an improved canny method for detecting color picture edges. Quaternion weighted average filter, Sobel vector gradient computation, interpolation-based non-maxima suppression, edge detection, and connection are the stages in the proposed approach. The suggested approach outperforms previous methods for detecting color picture edges in experiments, and it may be used to a wide range of color image processing tasks. Color image processing has become increasingly popular as a result of ubiquitous color picture applications such as image identification, video surveillance, and medical image analysis. Color image processing, on the other hand, is a more recent research than gray image processing. As a result, further study is necessary on this topic.
Study on color image processing based intelligent fruit sorting system
A computer vision gadget that may be used to propose automated high-speed fruit sorting. The fruit region was initially separated from the picture using an Otha-color-space dependent thresholding method; the blob algorithm was used to remove noises in the image; and the spline-interpolation based approach was used to identify fruit contour. The classification function in the fruit sorting process was chosen based on the color ratio of the fruit, which was obtained using the HSI color space. Fruit sorting was handled by the standard Bayes classifier, whose parameters were acquired using a research module. This approach was used to check crystal Fuji apples, and an average sorting accuracy of 90% was attained. Monochrome or greyscale pictures were used in the majority of primary systems. According to(Guo and Cao, 2004), color image processing systems have mostly been utilized in grading to assess fruit quality rather than to detect bruises. This study describes a color image processing-based vision system for sorting Fuji apples. The image-processing method first isolates the apple region from the image's backdrop, then calculates its color ratio, which is the most important factor in determining the quality of Fuji apples. A specific research module is introduced in the software, which is utilized to initialize the parameters of the traditional Bayes classifier responsible for apple classification.
Developing a framework for fruits detection from images
Several fruit identification methods have been developed based on color and form parameters. Colors and forms of diverse representations of objects, on the other hand, might be similar or identical. As a result, relying just on color and form features for fruit identification is insufficient. In light of this, the researcher offers in this study a technique for detecting things from pictures that employs color, form, and texture. They conducted many trials using fruit photos and determined that the gadget can virtually properly recognize fruit kinds from images. Fruit detection is a technique for identifying occurrences of fruits in photographs. The goal of fruit detection is to find all fruit representations. While there are only a few instances of the things in the photographs, their locations and sizes are vastly different. As a result, capturing fruit positions is a critical aspect of fruit identification from photos (Hossen et al., 2017).
An optimal color image edge detection approach
According to (Bora, 2018), identifying edges in color pictures is a relatively new issue in machine learning research. An effective technique for color picture edge detection is presented in this paper. The Canny edge detector is used in the proposed approach. Color calculations are carried out in the HSV color space. The shortcomings of the classic Canny edge detector are solved by replacing the Gaussian filter with an adaptive median filter, and a type-2 fuzzy set-based method is employed in the last phase of image segmentation for double threshold computation. The test results show that the proposed method is superior. Edges in an image may be described as the borders or outlines of the Selected area (ROI) contained in the picture, and they are the most significant component. It is the most widely used method in picture segmentation and the most fundamental stage in pattern recognition development. Image segmentation isolates different items in a picture, allowing image analysis to proceed more quickly. Segmentation algorithm based on discontinuity is called edge detection. The edge of an image function can be defined as a substantial discontinuity. The precision with which these discontinuities are identified and brought to light determines the efficiency of any edge detection approach. A gray scale edge is formed when the input image abruptly changes.
- Quote paper
- Randy Ramirez (Author), 2020, Trichoderma Classification System Based on Color Code Texture of Potato Dextrose Agar Solid Using Tensorflow with Prototype, Munich, GRIN Verlag, https://www.grin.com/document/1061354
Publish now - it's free