scholarly journals IDENTIFIKASI CITRA UKIRAN ORNAMEN TRADISIONAL BALI DENGAN METODE MULTILAYER PERCEPTRON

2021 ◽  
Vol 4 (1) ◽  
pp. 29-39
Author(s):  
I Gede Rusdy Mahayana Putra ◽  
Made Windu Antara Kesiman ◽  
Gede Aditra Pradnyana ◽  
I Made Dendi Maysanjaya
Keyword(s):  
Image Processing ◽  
Digital Image Processing ◽  
Digital Image ◽  
Multilayer Perceptron ◽  
Learning Algorithm ◽  
Gabor Filter ◽  
Confusion Matrix ◽  
Identification Accuracy ◽  
Average Accuracy ◽  
Fold Cross Validation

Balinese ornament carving are a cultural heritage that is owned by especially the Balinese people. However, especially Balinese people only know the shape of the carving without knowing the name and characteristics of the Balinese traditional carving ornaments. Based on these problems, the researchers have a solution to research about Balinese Ornament Carving Identification by utilizing digital image processing technology. In this study uses Gabor Filter as a feature extraction from the carved image that used and Multilayer Perceptron as a classifier. There are 18 (eighteen) classes of Balinese carving ornaments use in this study with a total of dataset is 268 (two hundred and sixty eight). The purpose of this study was to determine the level of identification  accuracy  of Balinese ornament carving with Multilayer Perceptron method. In the implementation using digital image processing technic with Multilayer Perceptron method was based on backpropagation learning algorithm with 10560 neuron input layers, 50 neuron hidden layers, and 18 neuron output layers as classifier obtained the accuracy for testing is 43%. Classification testing based on k-fold cross validation with K=5 results in average accuracy of 41.14% with optimum accuracy of 56% and accuracy testing with Confusion Matrix obtained the accuracy 43.3%, sensitivity 42.68% and specificity 96.87%. 

scholarly journals Machine Learning Techniques for Prediction of Lung Cancer

2020 ◽  
Vol 8 (6) ◽  
pp. 241-249
Keyword(s):  
Machine Learning ◽  
Lung Cancer ◽  
Data Mining ◽  
Image Processing ◽  
Digital Image Processing ◽  
Digital Image ◽  
Learning Algorithm ◽  
Confusion Matrix ◽  
Machine Learning Techniques ◽  
Image Dataset

Lung cancer has been one of the deadliest diseases in today’s decades. It has become one of the causes of death in both man and woman. There are various reasons for which lung cancer occurs but classification of tumor and predicting it in the right stage is the most important part. This paper focused on the numerous approaches has been derived for lung cancer detection from different literature survey to advance the ability of detection of cancer. Digital image processing and data mining both are equally important because for prediction either image dataset or statistical dataset is used so for pre-processing the image dataset digital image processing is applied for statistical dataset data mining is applied. After pre-processing, segmentation and feature extraction we apply various machine learning algorithm for the prediction of lung cancer. So first we have provided a sketch of Machine learning and then various fields like in image data or statistical data where machine learning has been used for classification. Once the classification is done confusion matrix is generated for calculating accuracy, sensitivity, precision, these method is used to measure the rate of accuracy of the proposed model.

scholarly journals Shrimps clusterization by size using digital image processing with CCA and DBSCAN

2020 ◽  
Vol 8 (2) ◽  
pp. 106-112
Author(s):  
Adri Priadana ◽  
Aris Wahyu Murdiyanto
Keyword(s):  
Image Processing ◽  
Light Intensity ◽  
Digital Image Processing ◽  
Digital Image ◽  
Spatial Clustering ◽  
Digital Camera ◽  
Harvest Time ◽  
Connected Component ◽  
Average Accuracy

The quality of farmed shrimps has several criteria, one of which is shrimp size. The shrimp selection was carried out by the contractor at the harvest time by grouping the shrimp based on their size. This study aims to apply digital image processing for shrimp clustering based on size using the connected component analysis (CCA) and density-based spatial clustering of applications with noise (DBSCAN) methods. Shrimp group images were taken with a digital camera at a light intensity of 1200-3200 lux. The clustering results were compared with clustering from direct observation by two experts, each of which obtained an accuracy of 79.81 % and 72.99 % so that the average accuracy of the method was 76.4 %.

scholarly journals Waste Collector Roboboat Using Neural Network Method Based on Tensorflow Framework

2021 ◽  
Vol 328 ◽  
pp. 02003
Author(s):  
Fiqhi I Achmad ◽  
Riza Alfita ◽  
Vivin N Rosida ◽  
Aji W Kunto ◽  
Hanifudin Sukri ◽  
...  
Keyword(s):  
Neural Network ◽  
Image Processing ◽  
Digital Image Processing ◽  
Digital Image ◽  
Navigation System ◽  
Research Result ◽  
Average Accuracy ◽  
Network Method ◽  
The Right ◽  
Main Factor

Today, the waste problem has become more serious, because the waste can cause environmental pollution and bad smell pollution. The less awareness of cleanliness is the main factor, especially the less awareness of throwing waste at the right place. Based on the data of the Ministry of Environment and Forestry “Environmental Ignorance Behavior” in 2008 said that around 72 percent of Indonesian less be aware of the waste problem generally with plastic waste, this waste will flow to the sea and make pollution. This research objective is to design and accomplish previous roboboat research that still has some drawbacks. This research employs digital image processing and neural network based on the tensor flow framework method to overcome less accurate waste detection as well as autopilot navigation system. The research result shows by using 3600 dataset images, the model has the lowest loss 0.9 and 64.3% average accuracy with various samples and distance.

A Hybrid Approach for Plant Leaf Disease Detection and Classification Using Digital Image Processing Methods

2020 ◽  
pp. 002072092095312
Author(s):  
Anusha Rao ◽  
S.B. Kulkarni
Keyword(s):  
Image Processing ◽  
Feature Extraction ◽  
Digital Image Processing ◽  
Digital Image ◽  
Gabor Filter ◽  
Hybrid Approach ◽  
Project Based Learning ◽  
Detection Accuracy ◽  
Plant Leaf ◽  
Leaf Disease

Detection of plant leaf disease has been considered an interesting research field which is helpful to improve the crop and fruit yield. Computer vision and machine learning based approaches have gained huge attraction in digital image processing field. Several visual computing based techniques have been presented in the past for early prediction of plant leaf diseases. However, detection accuracy is still considered as a challenging task. Hence, in order to overcome this issue, we introduce a novel hybrid approach carried out in three forms. During the first phase, image enhancement and image conversion scheme are incorporated, which helps to overcome the low-illumination and noise related issues. In the next phase, a combined feature extraction technique is developed by using GLCM, Complex Gabor filter, Curvelet and image moments. Finally, a Neuro-Fuzzy Logic classifier is trained with the extracted features. The proposed approach is implemented using MATLAB simulation tool where PlantVillage Database is considered for analysis. The average detection accuracy has been obtained as more than 90% for 2 test cases which shows that the proposed combination of feature extraction and image pre-processing process is able to obtain improved classification accuracy. This work is useful for the students of UG/PG programme to carry out Project-based learning.

scholarly journals Detection of Exudates and Microaneurysms in the Retina by Segmentation in Fundus Images

2021 ◽  
Author(s):  
◽  
E. Bernal-Catalán
Keyword(s):  
Image Processing ◽  
Diabetic Retinopathy ◽  
Digital Image Processing ◽  
Digital Image ◽  
Confusion Matrix ◽  
Fundus Images ◽  
Image Processing Techniques ◽  
Luminous Objects ◽  
Processing Techniques ◽  
The Ideal

This article proposes two methodologies for the detection of lesions in the retina, which may indicate the presence of diabetic retinopathy (DR). Through the use of digital image processing techniques, it is possible to isolate the pixels that correspond to a lesion of RD, to achieve segmenting microaneurysms, the edges of the objects contained in the image are highlighted in order to detect the contours of the objects to select by size those that meet an area of 15 to 25 pixels in the case of 512x512 images and identify the objects as possible microaneurysms, while for the detection of exudates the green channel is selected to contrast the luminous objects in the retinography and from the conversion to gray scale, a histogram is graphed to identify the ideal threshold for the segmentation of the pixels that belong to the exudates at the end of the optical disk previously identified by a specialist. A confusion matrix supervised by an ophthalmologist was created to quantify the results obtained by the two methodologies, obtaining a specificity of 0.94 and a sensitivity of 0.97, values that are outstanding to proceed with the classification stage.

scholarly journals Fruit Grading of Garcinia Binucao (Batuan) using Image Processing

2019 ◽  
Vol 8 (2) ◽  
pp. 1828-1832
Keyword(s):  
Image Processing ◽  
Computer Vision ◽  
Digital Image Processing ◽  
Digital Image ◽  
Color Features ◽  
Average Accuracy ◽  
Fruit Grading ◽  
Commercial Value ◽  
Light Green

Digital image processing, along with computer vision techniques, can be applied for automatic gradation of batuan fruits based on the quality of the fruit. It can increase the commercial value of the production. This paper presents an automated approach for fruit grading of Batuan fruits. First, it detects the batuan fruit and then after detection, it classifies the grade of the batuan fruit - Immature (light green), Mature (green) or Ripe (yellow). For gradation, we have used the color features and area of the fruit. The average accuracy for batuan fruit grading is up to 98%.

An Overview of Digital Image Processing and Recognition (Invited Paper)

1982 ◽  
Vol 40 ◽  
pp. 700-702
Author(s):  
R. C. Gonzalez
Keyword(s):  
Image Processing ◽  
New York ◽  
Digital Image Processing ◽  
Digital Image ◽  
Practical Implementation ◽  
Submarine Cable ◽  
Vigorous Growth ◽  
Digitized Pictures ◽  
Processing Techniques ◽  
And Storage

Interest in digital image processing techniques dates back to the early 1920's, when digitized pictures of world news events were first transmitted by submarine cable between New York and London. Applications of digital image processing concepts, however, did not become widespread until the middle 1960's, when third-generation digital computers began to offer the speed and storage capabilities required for practical implementation of image processing algorithms. Since then, this area has experienced vigorous growth, having been a subject of interdisciplinary research in fields ranging from engineering and computer science to biology, chemistry, and medicine.

A method to measure pores and fissures in geologic materials under S.E.M. by digital image processing

1978 ◽  
Vol 36 (1) ◽  
pp. 212-213
Author(s):  
L. Montoto ◽  
M. Montoto ◽  
A. Bel-Lan
Keyword(s):  
Image Processing ◽  
Optical Microscopy ◽  
Digital Image Processing ◽  
Digital Image ◽  
Digital Processing ◽  
Free Surfaces ◽  
Geologic Materials ◽  
Automatic Information ◽  
Detector Output ◽  
Rock Specimens

INTRODUCTION.- The physical properties of rock masses are greatly influenced by their internal discontinuities, like pores and fissures. So, these need to be measured as a basis for interpretation. To avoid the basic difficulties of measurement under optical microscopy and analogic image systems, the authors use S.E.M. and multiband digital image processing. In S.E.M., analog signal processing has been used to further image enhancement (1), but automatic information extraction can be achieved by simple digital processing of S.E.M. images (2). The use of multiband image would overcome difficulties such as artifacts introduced by the relative positions of sample and detector or the typicals encountered in optical microscopy.DIGITAL IMAGE PROCESSING.- The studied rock specimens were in the form of flat deformation-free surfaces observed under a Phillips SEM model 500. The SEM detector output signal was recorded in picture form in b&w negatives and digitized using a Perkin Elmer 1010 MP flat microdensitometer.

Digital image processing methods applied to the study of annealing time on semiconductor-metal eutectic composites

1988 ◽  
Vol 46 ◽  
pp. 848-849
Author(s):  
J. Hefter
Keyword(s):  
Image Processing ◽  
Digital Image Processing ◽  
Digital Image ◽  
Electronic Device ◽  
Processing System ◽  
Average Diameter ◽  
Eutectic Solidification ◽  
Metal Silicide ◽  
The Matrix ◽  
Microscopic Studies

Semiconductor-metal composites, formed by the eutectic solidification of silicon and a metal silicide have been under investigation for some time for a number of electronic device applications. This composite system is comprised of a silicon matrix containing extended metal-silicide rod-shaped structures aligned in parallel throughout the material. The average diameter of such a rod in a typical system is about 1 μm. Thus, characterization of the rod morphology by electron microscope methods is necessitated.The types of morphometric information that may be obtained from such microscopic studies coupled with image processing are (i) the area fraction of rods in the matrix, (ii) the average rod diameter, (iii) an average circularity (roundness), and (iv) the number density (Nd;rods/cm2). To acquire electron images of these materials, a digital image processing system (Tracor Northern 5500/5600) attached to a JEOL JXA-840 analytical SEM has been used.

Complementary Z-Contrast Imaging and Digital Image Processing

1985 ◽  
Vol 43 ◽  
pp. 304-305
Author(s):  
K. N. Colonna ◽  
G. Oliphant
Keyword(s):  
Image Processing ◽  
Heavy Metal ◽  
Digital Image Processing ◽  
Digital Image ◽  
Biological Tissue ◽  
Biological Imaging ◽  
Tissue Preparation ◽  
Contrast Imaging ◽  
Imaging Tool ◽  
Z Contrast

Harmonious use of Z-contrast imaging and digital image processing as an analytical imaging tool was developed and demonstrated in studying the elemental constitution of human and maturing rabbit spermatozoa. Due to its analog origin (Fig. 1), the Z-contrast image offers information unique to the science of biological imaging. Despite the information and distinct advantages it offers, the potential of Z-contrast imaging is extremely limited without the application of techniques of digital image processing. For the first time in biological imaging, this study demonstrates the tremendous potential involved in the complementary use of Z-contrast imaging and digital image processing.Imaging in the Z-contrast mode is powerful for three distinct reasons, the first of which involves tissue preparation. It affords biologists the opportunity to visualize biological tissue without the use of heavy metal fixatives and stains. For years biologists have used heavy metal components to compensate for the limited electron scattering properties of biological tissue.

Sign in / Sign up

Export Citation Format

Share Document