KMITL

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

Browse

Search Results

Now showing 1 - 10 of 23

Some of the metrics are blocked by your
consent settings
DIP-CBML: A New Classification of Thai Dragon Fruit Species from Images
(2023-01-01)
Yusamran, Naruwan
;
Hiransakolwong, Nualsawat
The attractiveness of dragon fruit is that it has a strange exterior, beautiful colors, and high nutritional value. In Thailand, there is both import and export of dragon fruit. Each package for export must contain only one species of dragon fruit. From the survey, there are seven species of dragon fruit cultivated in Thailand and only some farmers can identify them on his/her farm. Therefore, this research focuses on the classification of Thai dragon fruit from laboratory images and outdoor images; which is different from the previous works which studied only laboratory images. This method was named DIP-CBML that stands for digital image processing with content-based and machine learning. The method consists of image type identification, pre-processing, red and yellow classification, image background removal, and six classes of red species classification. From the results, DIP-CBML can work with both datasets. It gave 100%, 100% and 95.53% accuracy for the image type identification, red and yellow classification, and the classification of six red species respectively. Hopefully, this research will lead to the innovation for the pre-harvest classification of Thai dragon fruit cultivars, applied to industrial applications, and robot harvesting. In the future, may add value to the yield of Thai dragon fruit cultivation.
Some of the metrics are blocked by your
consent settings
CBML: CLASSIFICATION – THAI RED DRAGON FRUIT
(2022-11-01)
Yusamran, Naruwan
;
Hiransakolwong, Nualsawat
A dragon fruit is nutritious having low sugar levels suitable for diabetics. Its shape is spherical with petals surrounding the bark. In Thailand, 7 species of dragon fruit are grown. Its appearance can be divided into red bark (6 species) or yellow bark (1 species). People can separate between yellow and red bark. Since the red-skinned dragon fruit may have its fruit pulp in one of three possible colors: white, red or pink, it is difficult to guess from the outside what color of the fruit pulp will be. The big problem is that people cannot classify its species of red bark. Some people may be allergic or dislike dragon fruit in some species. Therefore, this research is trying to find the best way to automatically classify the species of Thai red dragon fruit from its image. The CBML stands for content base and Machine Learning. This method uses content base to extract key features (34 attributes), and uses Machine Learning for giving optimization results with the Support Vector Machine (SVM), setting kernel for polynomial in 8 degree. The results showed that the CBML method was able to identify any species of dragon fruits in the red bark group with an accuracy of 98.47%.
Some of the metrics are blocked by your
consent settings
DIPDEEP: Classification for Thai dragon fruit
(2022-02-22)
Yusamran, Naruwan
;
Hiransakolwong, Nualsawat
Thai dragon fruit is an interesting fruit with beautiful colors and high nutritional value, which can be used for food and pharmaceutical. In Thailand, there are 7 species of dragon fruits. Its skin can be divided into red or yellow groups, but inside can be divided into white, red or pink groups. If farmers know the species of dragon fruits, they can export fruits at a good price. Most people are unable to distinguish species of the dragon fruits. This paper is focus on classifying species of dragon fruits from images using digital image processing and deep learning called DIPDEEP. The DIPDEEP method has three steps; color space transformation, calculation ratio of the yellow and red color, classification using the deep learning method. First, the Thai dragon fruit images were classified into yellow (1 species) out from red (6 species). The Thai red dragon fruit was resized into 100x100 pixel resolutions in the pre-processing step only 6 species. Then, the red class was sent to classify again using the deep learning method. The experiments were processed in a dataset with 9,754 dragon fruit images on a black background (laboratory), and 10,072 images of dragon fruits at outdoor environment (outdoor). The results showed that accuracy of classification between the red and yellow dragon fruit for laboratory and outdoor datasets was 100% and 95.26%, respectively. The red dragon fruit is classified its species with accuracy 98.80%. The DIPDEEP has the smallest file size, and can save workload time because of separating yellow skin out at first step.
Some of the metrics are blocked by your
consent settings
Segmentation of overlapping characters in lanna using mixed algorithm
(2020-08-01)
Kosarat, Rujipan
;
Hiransakolwong, Nualsawat
Lanna language is a popular language in the northern part of Thailand. The segmentation of printed Lanna characters is a challenging problem. Normally, the segmentation method of characters begins with using horizontal and vertical histograms into line segmentation and character segmentation. Then, when written the output will be the correct clear characters, overlapping characters and touching characters. Frequently, there are some problems with the overlapping characters and the touching characters. This paper focuses on only the overlapping characters, overlapping between alphabets and vowels. The methods will segment the overlapping characters by histogram techniques, beginning with splitting, rotating and merging the characters. The experiments used the printed Lanna characters in many books as a data set. The results achieved an accuracy rate of 96.72%.
Some of the metrics are blocked by your
consent settings
EQ101: Emotional Quotient Assessment A Mobile Application for Students
(2020-02-18)
Hiransakolwong, Nualsawat
Emotional Quotient or EQ means being able to realize feelings of oneself and others to create self-motivation and can handle various emotions become good, smart, and happy. There are only web applications for EQ assessment, but it cannot record any EQ history of the tests for seeing personal development of EQ and cannot report the test results to teachers or parents. Nowadays, most students use smartphones. This research presents an application in EQ assessment for students aging 12-17 years with processing on smartphones using Android OS. There were randomly selected 31 students from a local school to test with this application. The experiment results show that the application can help these students to know their EQ assessment results that were reported to their parents and counselor teachers. The application can record the previous results also to see the development of EQ in personal. At the same time, the executive board of school can use the histogram of EQ assessment results to plan for school activities that are suitable for developing EQ to their students. This can help students avoid with depression. All users gave an excellent level of satisfaction with this application on an average of 4.88 from 5.
Some of the metrics are blocked by your
consent settings
Segmentation of touching character printed lanna script using junction point
(2018-10-01)
Kosarat, Rujipan
;
Hiransakolwong, Nualsawat
In the northern part of Thailand since 1802, Lanna characters were popular as ancient characters. The segmentation of printed documents in Lanna characters is a challenging problem, such as the partial overlapping of characters and touching characters. This paper focuses on only the touching characters such as touching between consonants and vowels. Segmentation method begins with the horizontal histogram and then vertical histogram for segmentation of text lines and characters, respectively. The results are characters consisted of correct clear characters, partial overlapping characters, and touching characters. The proposed method computes the left edge junction points and right edge junction points. Then find their maximum numbers and find the value of its row to separate consonant and vowel from touching. The trial over the text documents printed in Lanna characters can be processed with an accuracy of 95.81%.
Some of the metrics are blocked by your
consent settings
Automatic detection of exudates in retinal images based on threshold moving average models
(2015-03-20)
Wisaeng, Kittipol
;
Hiransakolwong, Nualsawat
;
Pothiruk, Ekkarat
Since exudates diagnostic procedures require the attention of an expert ophthalmologist, as well as regular monitoring of the disease and the workload of expert ophthalmologists will eventually exceed the current screening capabilities. Retinal imaging technology is a current practice screening capability provide a great potential solution. In this paper, a fast and robust automatic detection of exudates based on moving average histogram models of the fuzzy image, and then derives the better histogram. After segmentation of candidate exudates, the true exudates were prune based on Sobel edge detector and automatic Otsu’s thresholding algorithm is presented that results in the accurate location of the exudates in digital retinal images. To compare the performance of exudates detection methods we have constructed a large database of digital retinal images. The method was trained on a set of 200 retinal images, and tested on a completely independent set of 1 220 retinal images. Results show that the exudates detection method performs overall best sensitivity, specificity, and accuracy are 90.42, 94.60, and 93.69%, respectively.
Some of the metrics are blocked by your
consent settings
Identification of exudates using fuzzy mathematical morphology
(2014-01-01)
Wisaeng, Kittipol
;
Hiransakolwong, Nualsawat
;
Pothiruk, Ekkarat
Diabetic Retinopathy is the damage to the retina caused by complication and the most common cause of blindness in Thailand. Retinal image is essential for expert ophthalmologists to diagnose diseases. Several of method can achieve good performance on retinal feature are clearly visible. Unfortunately, the color retinal image in Thailand are low-resolution images. The existing method cannot identified lowresolution image. Therefore, this study is part of a larger effort to develop a new method for identification of exudates in low-resolution retinal image. In this study a fuzzy mathematical morphology based on fuzzy logical operator and mathematical morphology method is presented. The color retinal image are segmented by using fuzzy logical operator following key preprocessing step, i.e., color normalization, contrast enhancement, noise removal and color space selection. Afterward, a segmentation using mathematical morphology method was applied in this step. This enables its difference in our methods compared to other approach and the methods can achieve good performance even on lowresolution retinal images. Respect to the experimental results, the results obtained with fuzzy mathematical morphology better than the ones obtained with the fuzzy logical operator only method.
Some of the metrics are blocked by your
consent settings
Grid-line watermarking: A novel method for creating a high-performance text-image watermark
(2013-01-01)
Yawai, Wiyada
;
Hiransakolwong, Nualsawat
This study aims to discover an effective method to watermark any text image in any language. The ultimate goal of this work is to generate invisible and more robust watermarks, increase the hiding capacity, and identify any change in the original text. Generally these are the limitations of most text-image watermarking methods. Using a grid of horizontal and vertical lines is one of the most effective methods to overcome these limitations. These grid lines run across textimage character-skeleton lines to finely mark and detect watermarks on these line-character intersection points; there is one intersection for one specific zero watermark pixel. Each intersection point is defined by one hiding bit of the watermark such that the more reference horizontal and vertical lines of grid pattern are plotted, the more points of intersection are obtained. This approach increases the hiding-bit capacity to watermark embedding, which is a disadvantage shared by many text-image watermarking methods. In addition, if all positions of all intersection points are collected, these points will act as the identity reference points of all original characters to verify their integrities after they are modified.
Some of the metrics are blocked by your
consent settings
Noise filtering in unsupervised clustering using computation intelligence
(2012-12-01)
Lowongtrakool, Chaloemchai
;
Hiransakolwong, Nualsawat
The present data used for clustering contains data record not related to processing. This may be spam or noise causing error and delay of processing. However, it is necessary to process this kind of data together but the results may be incorrect, depending on the quantity of noise. Therefore, it will be much better if data cleaning is conducted before processing system data. This paper proposes a method to develop unsupervised clustering intelligence to reduce the quantity of spam. This computational intelligence system applies the first layer of radial basis function network as an input layer of the system for incremental work. The results of system can provide level of accuracy of least related membership in any cluster which is called the data not relevant to the dataset or the data not associated with the dataset. According to an experiment, the data from UCI machine learning repository was used to test the system efficiency while classification algorithm in a program called weka3.6 was also utilized to test the system accuracy. The results from noise filtering help the data processing more precise, compared to the processing without the noise filtering.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results