KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 10

Some of the metrics are blocked by your
consent settings
Using a Computer Vision System for Monitoring the Exterior Characteristics of Damaged Apples
(2025-10-01)
Al-Riyami, Zamzam
;
Al-Dairi, Mai
;
Pathare, Pankaj B.
;
Kramchote, Somsak
Mechanical damage like bruises produced during postharvest handling can lower market value, affect nutritional value, and pose food safety risks. The study evaluated bruises on apples using image processing. This research focuses on using computer vision for apple fruit damage detection. The fruits were subjected to three levels of impact using three ball weights (66, 98, and 110 g) dropped from 50 cm height and stored at 22 °C. The overall impact energies generated were 0.323 J (low), 0.480 J (medium), and 0.539 J (high). The bruise area and susceptibility of the damage, surface area of the fruit, and color were measured manually (colorimeter) and by image processing. The study found that the bruise area was significantly affected by impact force, where 110 g (0.539 J) damaged apples showed a bruise area of 4.24 cm<sup>2</sup> after 21 days of storage at 22 °C. The images showed a significant change in the RGB values (Red, Green, Blue) over 21 days of storage when impacted at 0.539 J. The study showed that the greater the impact energy effect, the higher the weight loss under constant conditions of storage. After 21 days of storage, the 110 g mechanically damaged apples recorded the highest percentage of weight loss (6.362%). The study found a significant decrease in the surface area of 110 g bruised apples, with a smaller decrease in surface area for 66 g bruised fruit. The use of computer vision to detect bruise damage and other quality attributes of Granny Smith apples can be highly recommended to detect their losses.
Some of the metrics are blocked by your
consent settings
Estimation of the Weight and Volume of Lime (Citrus aurantifolia (Christm.) Swingle) Fruit Using Computer Vision Based on Traditional Machine Learning and Deep Learning
(2024-10-01)
Onmankhong, Jiraporn
;
Poonpakdee, Pasu
;
Lapcharoensuk, Ravipat
The post-harvest process is important to increasing the market value of limes and requires focus. During this process, limes are graded and categorized based on size, weight, and volume. Therefore, identifying efficient means of estimating these properties is very important and remains an open research area. This study applies the concept of computer vision based on traditional machine learning algorithms (partial least square regression (PLS), epsilon-support vector regression (ε-SVR), decision tree (DT), random forest (RF), adaptive boosting (AB), gradient boosting (GB), Bagging meta-estimator (BME), and extremely randomized trees (ERTs)) and pre-trained deep learning (InceptionV3, MoblieNetV2, ResNet50, and VGG-16) for estimating the weight and volume of limes. Our findings showed that the BME and ResNet50 could yield the highest performance for estimating the weight and volume of limes. The BME produced (Formula presented.) values of 0.954 and 0.882 for weight and volume, respectively, while the (Formula presented.) values of ResNet50 models were between 0.951 and 0.957 for weight and volume, respectively. This study concluded that computer vision based on both traditional machine learning and deep learning could be used to estimate the weight and volume of limes. The approach proposed in this study can be adopted for applications related to computer vision in the post-harvest process.
Some of the metrics are blocked by your
consent settings
Hybrid Deep Learning and FAST-BRISK 3D Object Detection Technique for Bin-picking Application
(2024-01-01)
Taweesoontorn, Thanakrit
;
Yanyong, Sarucha
;
Konghuayrob, Poom
In the field of industrial robotics, robotic arms have been significantly integrated, driven by their precise functionality and operational efficiency. We here propose a hybrid method for bin-picking tasks using a collaborative robot, or cobot combining the You Only Look Once version 5 (YOLOv5) convolutional neural network (CNN) model for object detection and pose estimation with traditional feature detection based on the features from accelerated segment test (FAST) technique, feature description using binary robust invariant scalable keypoints (BRISK) algorithms, and matching algorithms. By integrating these algorithms and utilizing a low-cost depth sensor camera for capturing depth and RGB images, the system enhances real-time object detection and pose estimation speed, facilitating accurate object manipulation by the robotic arm. Furthermore, the proposed method is implemented within the robot operating system (ROS) framework to provide a seamless platform for robotic control and integration. We compared our results with those of other methodologies, highlighting the superior object detection accuracy and processing speed of our hybrid approach. This integration of robotic arm, camera, and AI technology contributes to the development of industrial robotics, opening up new possibilities for automating challenging tasks and improving overall operational efficiency.
Some of the metrics are blocked by your
consent settings
The current state of artificial intelligence-augmented digitized neurocognitive screening test
(2023-01-01)
Sirilertmekasakul, Chananchida
;
Rattanawong, Wanakorn
;
Gongvatana, Assawin
;
Srikiatkhachorn, Anan
The cognitive screening test is a brief cognitive examination that could be easily performed in a clinical setting. However, one of the main drawbacks of this test was that only a paper-based version was available, which restricts the test to be manually administered and graded by medical personnel at the health centers. The main solution to these problems was to develop a potential remote assessment for screening individuals with cognitive impairment. Currently, multiple studies have been adopting artificial intelligence (AI) technology into these tests, evolving the conventional paper-based neurocognitive test into a digitized AI-assisted neurocognitive test. These studies provided credible evidence of the potential of AI-augmented cognitive screening tests to be better and provided the framework for future studies to further improve the implementation of AI technology in the cognitive screening test. The objective of this review article is to discuss different types of AI used in digitized cognitive screening tests and their advantages and disadvantages.
Some of the metrics are blocked by your
consent settings
DEVELOPMENT OF OBJECT DETECTION AND CLASSIFICATION WITH YOLOV4 FOR SIMILAR AND STRUCTURAL DEFORMED FISH
(2022-03-31)
Kuswantori, Ari
;
Suesut, Taweepol
;
Tangsrirat, Worapong
;
Nunak, Navaphattra
Food scarcity is an issue of concern due to the continued growth of the human population and the threat of global warming and climate change. Increasing food production is expected to meet the challenges of food needs that will continue to increase in the future. Automation is one of the solutions to increase food productivity, including in the aquaculture industry, where fish recognition is essential to support it. This paper presents fish recognition using YOLO version 4 (YOLOv4) on the «Fish-Pak» dataset, which contains six species of identical and structurally damaged fish, both of which are characteristics of fish processed in the aquaculture industry. Data augmentation was generated to meet the validation criteria and improve the data balance between classes. For fish images on a conveyor, flip, rotation, and translation augmentation techniques are appropriate. YOLOv4 was applied to the whole fish body and then combined with several techniques to determine the impact on the accuracy of the results. These techniques include landmarking, subclassing, adding scale data, adding head data, and class elimination. Performance for each model was evaluated with a confusion matrix, and analysis of the impact of the combination of these techniques was also reviewed. From the experimental test results, the accuracy of YOLOv4 for the whole fish body is only 43.01 %. The result rose to 72.65 % with the landmarking technique, then rose to 76.64 % with the subclassing technique, and finally rose to 77.42 % by adding scale data. The accuracy did not improve to 76.47 % by adding head data, and the accuracy rose to 98.75 % with the class elimination technique. The final result was excellent and acceptable.
Some of the metrics are blocked by your
consent settings
Fish Recognition Optimization in Various Backgrounds Using Landmarking Technique and YOLOv4
(2022-01-01)
Kuswantori, Ari
;
Suesut, Taweepol
;
Tangsrirat, Worapong
;
Satthamsakul, Sutham
The identification and categorization of fish is a popular and fascinating research topic. Many researchers have developed expertise in fish detection, both underwater and outside the water, which is particularly beneficial for population management and aquaculture. This paper proposes a fish recognition approach using the landmarking methodology with YOLO version 4 to identify and categorize fish with different backdrop circumstances. The approach can be used both underwater and on land. The proposed approach was evaluated using four distinct types of fish from the BYU dataset. The final test result determined that the accuracy reached 96.60%, with an average classification score of 99.67% at the 60% threshold. The result is 4.94 % better than the most frequent traditional labelling approach.
Some of the metrics are blocked by your
consent settings
Turbidity of Coconut Oil Determination Using the MAMoH Method in Image Processing
(2021-01-01)
Palananda, Attapon
;
Kimpan, Warangkhana
In general, considering standard production, as well as coconut oil production, in oil consumption industries is an important factor. Oil color is an important element, as it is an important factor for consumers or buyers in selecting coconut oil. In the process of producing coconut oil, the cold-pressed method has been chosen to maintain the essential quality of coconut oil. The quality of the coconut oil is inspected from the production process by means of light passing through the coconut oil. Then, the production staff compares the turbidity of coconut oil with the master sample. The turbidity of coconut oil in every production must be compared with a master sample to maintain standards control. According to previous studies, there are many methods for determining coconut oil turbidity. One method that has been utilized is determining turbidity from light passing through the medium in which the transmitted light can be absorbed through the turbidity of the variable medium. This process is applied together with image processing to determine the coconut oil turbidity. In this research, we propose a method for measuring coconut oil turbidity by the Moving Average Median of Hue (MAMoH), which is better in detecting the coconut oil turbidity than the Median of Gray Scale (MoGS) method, Median of Hue (MoH) method, and Random Position Median of Hue (RPMoH) method. In terms of the percentage accuracy of the efficiency test; the MAMoH method has 99 percent accuracy, while the MoGS method is not applicable, the MoH method has 88.04 percent accuracy, and the RPMoH method has 85.91 percent accuracy. Thus, the MAMoH method is considered an appropriate method for measuring coconut oil turbidity.
Some of the metrics are blocked by your
consent settings
Defect detection of particleboards by visual analysis and machine learning
(2019-07-01)
Prasitmeeboon, Pitcha
;
Yau, Henry
Particleboards may exhibit several defect types caused by a variety of sources during the manufacturing process. It is essential to quickly determine when a defect is present and localize the fault so that the board can either be fixed or discarded. Several methods have been already been developed to address this issue to varying degrees of success. In this work, a novel process is presented which quickly determines whether a defect exists or not using traditional machine learning techniques on a bivariate color histogram of the particleboard and then localize the defect using automated image manipulation techniques. The workflow of quickly determining if a defect is present then using a more computationally intensive technique to localize and classify the defect can be extended to use other methods or even to other processes.
Some of the metrics are blocked by your
consent settings
A computer vision based vehicle detection and counting system
(2016-03-23)
Seenouvong, Nilakorn
;
Watchareeruetai, Ukrit
;
Nuthong, Chaiwat
;
Khongsomboon, Khamphong
;
Ohnishi, Noboru
A vehicle detection and counting system plays an important role in an intelligent transportation system, especially for traffic management. This paper proposes a video-based method for vehicle detection and counting system based on computer vision technology. The proposed method uses background subtraction technique to find foreground objects in a video sequence. In order to detect moving vehicles more accurately, several computer vision techniques, including thresholding, hole filling and adaptive morphology operations, are then applied. Finally, vehicle counting is done by using a virtual detection zone. Experimental results show that the accuracy of the proposed vehicle counting system is around 96%.
Some of the metrics are blocked by your
consent settings
Near point light source location estimation from shadow edge correspondence
(2015-09-23)
Chotikakamthorn, Nopporn
The problem of near point light source location estimation is considered. The approach described here is based on a relationship between the cast shadow edge, the shadow-casting surface, and the light source. Given an optical-depth image pair available from an RGB-D camera, a method for near point light source location estimation is described. The method can be applied to a shadow-casting object with arbitrary and unknown geometry. Exhaustive multidimensional search is avoided by the proposed iterative one-dimensional search algorithm. Experimental results with real images are included.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results