Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis
Nowadays, despite a negative impact on the natural environment, coal combustion is still a significant energy source. One way to minimize the adverse side effects is sophisticated combustion technologies, such as, e.g., staged combustion, co-combustion with biomass, and oxy-combustion. Maintaining t...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/2/500 |
| _version_ | 1797412594822152192 |
|---|---|
| author | Zbigniew Omiotek Andrzej Kotyra |
| author_facet | Zbigniew Omiotek Andrzej Kotyra |
| author_sort | Zbigniew Omiotek |
| collection | DOAJ |
| description | Nowadays, despite a negative impact on the natural environment, coal combustion is still a significant energy source. One way to minimize the adverse side effects is sophisticated combustion technologies, such as, e.g., staged combustion, co-combustion with biomass, and oxy-combustion. Maintaining the combustion process at its optimal state, considering the emission of harmful substances, safe operation, and costs requires immediate information about the process. Flame image is a primary source of data which proper processing make keeping the combustion at desired conditions, possible. The paper presents a method combining flame image processing with a deep convolutional neural network (DCNN) that ensures high accuracy of identifying undesired combustion states. The method is based on the adaptive selection of the gamma correction coefficient (<i>G</i>) in the flame segmentation process. It uses the empirically determined relationship between the <i>G</i> coefficient and the average intensity of the R image component. The pre-trained VGG16 model for classification was used. It provided accuracy in detecting particular combustion states on the ranging from 82 to 98%. High accuracy and fast processing time make the proposed method possible to apply in the real systems. |
| first_indexed | 2024-03-09T05:04:23Z |
| format | Article |
| id | doaj.art-bee4efeb81c84f5182a33ce453ad1a00 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T05:04:23Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-bee4efeb81c84f5182a33ce453ad1a002023-12-03T12:56:29ZengMDPI AGSensors1424-82202021-01-0121250010.3390/s21020500Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion DiagnosisZbigniew Omiotek0Andrzej Kotyra1Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, 20-618 Lublin, PolandFaculty of Electrical Engineering and Computer Science, Lublin University of Technology, 20-618 Lublin, PolandNowadays, despite a negative impact on the natural environment, coal combustion is still a significant energy source. One way to minimize the adverse side effects is sophisticated combustion technologies, such as, e.g., staged combustion, co-combustion with biomass, and oxy-combustion. Maintaining the combustion process at its optimal state, considering the emission of harmful substances, safe operation, and costs requires immediate information about the process. Flame image is a primary source of data which proper processing make keeping the combustion at desired conditions, possible. The paper presents a method combining flame image processing with a deep convolutional neural network (DCNN) that ensures high accuracy of identifying undesired combustion states. The method is based on the adaptive selection of the gamma correction coefficient (<i>G</i>) in the flame segmentation process. It uses the empirically determined relationship between the <i>G</i> coefficient and the average intensity of the R image component. The pre-trained VGG16 model for classification was used. It provided accuracy in detecting particular combustion states on the ranging from 82 to 98%. High accuracy and fast processing time make the proposed method possible to apply in the real systems.https://www.mdpi.com/1424-8220/21/2/500image processingflame segmentationclassificationdeep learningVGG16convolutional neural networks |
| spellingShingle | Zbigniew Omiotek Andrzej Kotyra Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis Sensors image processing flame segmentation classification deep learning VGG16 convolutional neural networks |
| title | Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis |
| title_full | Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis |
| title_fullStr | Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis |
| title_full_unstemmed | Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis |
| title_short | Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis |
| title_sort | flame image processing and classification using a pre trained vgg16 model in combustion diagnosis |
| topic | image processing flame segmentation classification deep learning VGG16 convolutional neural networks |
| url | https://www.mdpi.com/1424-8220/21/2/500 |
| work_keys_str_mv | AT zbigniewomiotek flameimageprocessingandclassificationusingapretrainedvgg16modelincombustiondiagnosis AT andrzejkotyra flameimageprocessingandclassificationusingapretrainedvgg16modelincombustiondiagnosis |