![TELKOMNIKA Indonesian Journal of Electrical Engineering
Vol. 14, No. 2, May 2015, pp. 318 ~ 322
DOI: 10.11591/telkomnika.v14i2.7725 318
Received January 27, 2015; Revised March 23, 2015; Accepted April 17, 2015
The Detection of Straight and Slant Wood Fiber through
Slop Angle Fiber Feature
Ratri Dwi Atmaja*, Erwin Susanto, Junartho Halomoan, Gurnita K I, Muhammad Ary Murti
School of Electrical Engineering, Telkom University
Jalan Telekomunikasi no.1, Terusan Buah Batu, Bandung 40257, Indonesia
*Corresponding author, e-mail: ratridwiatmaja@telkomuniversity.ac.id
Abstract
Quality control is one of important process that can not be avoided in industry. Image processing
technique is required to distinguish the quality of wood. If it can be done automatically by the computer, it
will be very helpful. This paper discusses the detection of straight and slant wood fiber to distinguish its
quality. This paper proposes an algorithm by using only two features i.e. mean (average value of slop
angle fiber) and maximumangle (the maximum value of slop angle fiber). Then the classification method is
used by tresholding. The result shows the performance is achieved on accuracy 79.2%
Keywords: detection, slop angle fiber feature, algorithm of feature extraction
Copyright © 2015 Institute of Advanced Engineering and Science. All rights reserved.
1. Introduction
Image classification technique requires a fairly long step. It starts from image
segmentation, object identification, feature extraction, feature selection and classification [5].
Wood image real-time segmentation algorithm based on video processing has been proposed
by Ratri [1] and has achieved 100% accuracy. This paper is a continuation of the research
paper [1] to distinguish the straight and slant fiber. The samples are taken from previous studies
on paper [1] and this paper only focus to the feature extraction algorithm.
2. Research Method
Figure 1 is the flowchart of algorithm proposed. The samples are taken using webcame
and IP camera. Bwareaopen is used to eliminate the noise (small objects) in binary image. The
“bwlabel” is used to find connected objects in binary image. Example of binary image taken and
output “bwlabel” results is shown in Figure 2.
Figure 1. Flowchart of algorithm proposed](https://image.slidesharecdn.com/1823mar157725paperratridwiatmajaeditproofed-171220081740/85/The-Detection-of-Straight-and-Slant-Wood-Fiber-through-Slop-Angle-Fiber-Feature-1-320.jpg)
![TELKOMNIKA ISSN: 2302-4046
The Detection of Straight and Slant Wood Fiber Through Slop Angle Fiber… (Ratri Dwi Atmaja)
319
(a) (b)
Figure 2. (a) Example of binary image taken and (b) Output bwlabel results from Figure 2(a)
Figure 3. Flowchart to find angle and data
N from Figure 3 is the number of connected objects in bwlabel result. From Figure 2(b),
there are 3 connected objects so it get N=3. Matrix is binary image of a connected object. There
are 3 matrixs on Figure 2(b).
Figure 4. Third matrix from Figure 2(b)
Area is the amount of 1 valued pixel in matrix. It is obtained area=3 from Figure 4. Then,
A is 1-dimensional matrix which is represent the row position of 1 valued pixel taken from
matrix. while B is 1-dimensional matrix which is represent the column position of 1 valued pixel
taken from matrix. From Figure 4, it is obtained A=[2 1 2] then B=[4 5 5]. It means that there are
3 pixels of 1 valued pixel in coordinate (2,4), (1,5), and (2,5). Min is the lowest value of A, while
max is the highest value of A. If A=[2 1 2] then obtained min=1 and max=2.](https://image.slidesharecdn.com/1823mar157725paperratridwiatmajaeditproofed-171220081740/85/The-Detection-of-Straight-and-Slant-Wood-Fiber-through-Slop-Angle-Fiber-Feature-2-320.jpg)
![ ISSN: 2302-4046
TELKOMNIKA Vol. 14, No. 2, May 2015 : 318 – 322
320
(a) (b)
Figure 5. (a) Flowchart to find leftrow and leftcolumn and (b) Flowchart to find rightrow and
rightcolumn
length is the length of A, If A=[2 1 2] so it is obtained length =3. Then, angle is an
object slope angle (wood fiber) and can be found with this equation:
tan
| |
| |
Data is a 1-dimensional matrix contains the values of angle at a wood image. Mean is
the average of angle value while maximumangle is the highest angle value. These mean and
maximumangle are used as feature vector. Whereas, the classification uses tresholding
method. Tresholding is done by the following rules:
a) If maximumangle < x or mean < y, It is decided as straight fiber
b) If maximumangle >= x and mean >= y, It is decided as slant fiber
3. Results and Analysis
The samples which are used are taken from wood processing industry with the size
20cm x 8cm. Figure 6 is the sample:
(a) (b)
Figure 6. (a) Sample taken through webcame and (b) Sample taken through IP camera](https://image.slidesharecdn.com/1823mar157725paperratridwiatmajaeditproofed-171220081740/85/The-Detection-of-Straight-and-Slant-Wood-Fiber-through-Slop-Angle-Fiber-Feature-3-320.jpg)
![TELKOMNIKA ISSN: 2302-4046
The Detection of Straight and Slant Wood Fiber Through Slop Angle Fiber… (Ratri Dwi Atmaja)
321
To see the performance of the proposed algorithm, it is tested by using the three
scenarios:
a) The samples are taken by using two types of cameras to find out which is the best.
b) Do the optimization of x to find out the best value of x
c) Do the optimization of y to find out the best value of y
Table 1. The result of first scenario
No Type of Camera
Number of samples
Total Accuracy (%)
Straight fiber Slant fiber
1 Webcam 592 woods 356 woods 948 woods 71.73
2 IP Camera 150 woods 100 woods 250 woods 77.2
The experiments in Table 1 is done with a value of x = 4 and y = 3.2 then, the results
show the accuracy using IP camera is higher than webcame. These results are used as a
reference for using the sample of IP camera in the next scenario.
Table 2. The result of second scenario
No Value of x (degrees) Accuracy (%)
1 2 77.2
2 3 77.2
3 4 77.2
4 10 77.6
5 15 74
6 20 66.8
The experiments in Table 2 is done with a value of y = 3.2 then, the results show that
optimal value is on the value of x = 10. In other words, straight wood fiber has a value of slope
angle fiber below 10 degrees, and for slant fiber wood has a value of slop angle fiber began
over 10 degrees.
Table 3. The result of third scenario
No Value of y (degrees) Accuracy (%)
1 2.2 68.4
2 3.2 77.6
3 3.7 78.4
4 4 79.2
5 4.2 78.4
6 4.7 74.8
7 5.2 72
The experiments of table 3 is done with a value of x = 10 then, the results show that
optimal value is on the value of y = 4. In other words, straight wood fiber has an average value
of slop angle fiber below 4 degrees, and for slant fiber wood has an average value of slop angle
fiber began over 4 degrees.
4. Conclusion
This research has an algorithm with 79.2% accuracy by using only two features i.e.
mean (average value of slop angle fiber) and maximumangle (the maximum value of slop angle
fiber). This algorithm can be adapted to other cases that have a same typical object.
References
[1] Ratri Dwi Atmaja. Wood image real-time segmentation algorithm based on video processing.
International Journal of Imaging & Robotics. 2014; 15(1).](https://image.slidesharecdn.com/1823mar157725paperratridwiatmajaeditproofed-171220081740/85/The-Detection-of-Straight-and-Slant-Wood-Fiber-through-Slop-Angle-Fiber-Feature-4-320.jpg)
![ ISSN: 2302-4046
TELKOMNIKA Vol. 14, No. 2, May 2015 : 318 – 322
322
[2] Haralick, Robert M, Linda G Shapiro. Computer and Robot Vision Volume I. Addison-Wesley. 1992:
28-48.
[3] FS Najafabadi, H Pourghassem. Surface and Corner Defect Detection on Tile Images Using Gabor
Features, Level Set Segmentation and Dot Product. International Journal of Imaging & Robotics.
2012; 8(2).
[4] Ahmad Nazri Ali, Mohd Zaid Abdullah. One Dimensional With Dynamic Features Vector For Iris
Classification Using Traditional Support Vector Machines. Journal of Theoretical and Applied
Information Technology. 2014; 70(1).
[5] Nur Shazwani Kamarudin, et al. Comparison Of Image Classification Techniques Using Caltech 101
Dataset. Journal of Theoretical and Applied Information Technology. 2015; 71(1).
[6] Nadeem Mahmood, et al. Image Segmentation Methods and Edge Detection: An Application To Knee
Joint Articular Cartilage Edge Detection. Journal of Theoretical and Applied Information Technology.
2015; 71(1).
[7] Ning Chen, Xiao-ping Song, Yi Liu. Edge Detection Based on Biomimetic Pattern Recognition.
TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(9).
[8] Qu zhongshui. An Algorithm of Image Quality Assessment Based on Data Fitting of Image Histogram.
TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(1).
[9] Hong-an Li, Jie Zhang, Baosheng Kang. Image Deformation Based on Wavelet Filter and Control
Curves. TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(5).](https://image.slidesharecdn.com/1823mar157725paperratridwiatmajaeditproofed-171220081740/85/The-Detection-of-Straight-and-Slant-Wood-Fiber-through-Slop-Angle-Fiber-Feature-5-320.jpg)
The Detection of Straight and Slant Wood Fiber through Slop Angle Fiber Feature
- 1. TELKOMNIKA Indonesian Journal of Electrical Engineering Vol. 14, No. 2, May 2015, pp. 318 ~ 322 DOI: 10.11591/telkomnika.v14i2.7725 318 Received January 27, 2015; Revised March 23, 2015; Accepted April 17, 2015 The Detection of Straight and Slant Wood Fiber through Slop Angle Fiber Feature Ratri Dwi Atmaja*, Erwin Susanto, Junartho Halomoan, Gurnita K I, Muhammad Ary Murti School of Electrical Engineering, Telkom University Jalan Telekomunikasi no.1, Terusan Buah Batu, Bandung 40257, Indonesia *Corresponding author, e-mail: [email protected] Abstract Quality control is one of important process that can not be avoided in industry. Image processing technique is required to distinguish the quality of wood. If it can be done automatically by the computer, it will be very helpful. This paper discusses the detection of straight and slant wood fiber to distinguish its quality. This paper proposes an algorithm by using only two features i.e. mean (average value of slop angle fiber) and maximumangle (the maximum value of slop angle fiber). Then the classification method is used by tresholding. The result shows the performance is achieved on accuracy 79.2% Keywords: detection, slop angle fiber feature, algorithm of feature extraction Copyright © 2015 Institute of Advanced Engineering and Science. All rights reserved. 1. Introduction Image classification technique requires a fairly long step. It starts from image segmentation, object identification, feature extraction, feature selection and classification [5]. Wood image real-time segmentation algorithm based on video processing has been proposed by Ratri [1] and has achieved 100% accuracy. This paper is a continuation of the research paper [1] to distinguish the straight and slant fiber. The samples are taken from previous studies on paper [1] and this paper only focus to the feature extraction algorithm. 2. Research Method Figure 1 is the flowchart of algorithm proposed. The samples are taken using webcame and IP camera. Bwareaopen is used to eliminate the noise (small objects) in binary image. The “bwlabel” is used to find connected objects in binary image. Example of binary image taken and output “bwlabel” results is shown in Figure 2. Figure 1. Flowchart of algorithm proposed
- 2. TELKOMNIKA ISSN: 2302-4046 The Detection of Straight and Slant Wood Fiber Through Slop Angle Fiber… (Ratri Dwi Atmaja) 319 (a) (b) Figure 2. (a) Example of binary image taken and (b) Output bwlabel results from Figure 2(a) Figure 3. Flowchart to find angle and data N from Figure 3 is the number of connected objects in bwlabel result. From Figure 2(b), there are 3 connected objects so it get N=3. Matrix is binary image of a connected object. There are 3 matrixs on Figure 2(b). Figure 4. Third matrix from Figure 2(b) Area is the amount of 1 valued pixel in matrix. It is obtained area=3 from Figure 4. Then, A is 1-dimensional matrix which is represent the row position of 1 valued pixel taken from matrix. while B is 1-dimensional matrix which is represent the column position of 1 valued pixel taken from matrix. From Figure 4, it is obtained A=[2 1 2] then B=[4 5 5]. It means that there are 3 pixels of 1 valued pixel in coordinate (2,4), (1,5), and (2,5). Min is the lowest value of A, while max is the highest value of A. If A=[2 1 2] then obtained min=1 and max=2.
- 3. ISSN: 2302-4046 TELKOMNIKA Vol. 14, No. 2, May 2015 : 318 – 322 320 (a) (b) Figure 5. (a) Flowchart to find leftrow and leftcolumn and (b) Flowchart to find rightrow and rightcolumn length is the length of A, If A=[2 1 2] so it is obtained length =3. Then, angle is an object slope angle (wood fiber) and can be found with this equation: tan | | | | Data is a 1-dimensional matrix contains the values of angle at a wood image. Mean is the average of angle value while maximumangle is the highest angle value. These mean and maximumangle are used as feature vector. Whereas, the classification uses tresholding method. Tresholding is done by the following rules: a) If maximumangle < x or mean < y, It is decided as straight fiber b) If maximumangle >= x and mean >= y, It is decided as slant fiber 3. Results and Analysis The samples which are used are taken from wood processing industry with the size 20cm x 8cm. Figure 6 is the sample: (a) (b) Figure 6. (a) Sample taken through webcame and (b) Sample taken through IP camera
- 4. TELKOMNIKA ISSN: 2302-4046 The Detection of Straight and Slant Wood Fiber Through Slop Angle Fiber… (Ratri Dwi Atmaja) 321 To see the performance of the proposed algorithm, it is tested by using the three scenarios: a) The samples are taken by using two types of cameras to find out which is the best. b) Do the optimization of x to find out the best value of x c) Do the optimization of y to find out the best value of y Table 1. The result of first scenario No Type of Camera Number of samples Total Accuracy (%) Straight fiber Slant fiber 1 Webcam 592 woods 356 woods 948 woods 71.73 2 IP Camera 150 woods 100 woods 250 woods 77.2 The experiments in Table 1 is done with a value of x = 4 and y = 3.2 then, the results show the accuracy using IP camera is higher than webcame. These results are used as a reference for using the sample of IP camera in the next scenario. Table 2. The result of second scenario No Value of x (degrees) Accuracy (%) 1 2 77.2 2 3 77.2 3 4 77.2 4 10 77.6 5 15 74 6 20 66.8 The experiments in Table 2 is done with a value of y = 3.2 then, the results show that optimal value is on the value of x = 10. In other words, straight wood fiber has a value of slope angle fiber below 10 degrees, and for slant fiber wood has a value of slop angle fiber began over 10 degrees. Table 3. The result of third scenario No Value of y (degrees) Accuracy (%) 1 2.2 68.4 2 3.2 77.6 3 3.7 78.4 4 4 79.2 5 4.2 78.4 6 4.7 74.8 7 5.2 72 The experiments of table 3 is done with a value of x = 10 then, the results show that optimal value is on the value of y = 4. In other words, straight wood fiber has an average value of slop angle fiber below 4 degrees, and for slant fiber wood has an average value of slop angle fiber began over 4 degrees. 4. Conclusion This research has an algorithm with 79.2% accuracy by using only two features i.e. mean (average value of slop angle fiber) and maximumangle (the maximum value of slop angle fiber). This algorithm can be adapted to other cases that have a same typical object. References [1] Ratri Dwi Atmaja. Wood image real-time segmentation algorithm based on video processing. International Journal of Imaging & Robotics. 2014; 15(1).
- 5. ISSN: 2302-4046 TELKOMNIKA Vol. 14, No. 2, May 2015 : 318 – 322 322 [2] Haralick, Robert M, Linda G Shapiro. Computer and Robot Vision Volume I. Addison-Wesley. 1992: 28-48. [3] FS Najafabadi, H Pourghassem. Surface and Corner Defect Detection on Tile Images Using Gabor Features, Level Set Segmentation and Dot Product. International Journal of Imaging & Robotics. 2012; 8(2). [4] Ahmad Nazri Ali, Mohd Zaid Abdullah. One Dimensional With Dynamic Features Vector For Iris Classification Using Traditional Support Vector Machines. Journal of Theoretical and Applied Information Technology. 2014; 70(1). [5] Nur Shazwani Kamarudin, et al. Comparison Of Image Classification Techniques Using Caltech 101 Dataset. Journal of Theoretical and Applied Information Technology. 2015; 71(1). [6] Nadeem Mahmood, et al. Image Segmentation Methods and Edge Detection: An Application To Knee Joint Articular Cartilage Edge Detection. Journal of Theoretical and Applied Information Technology. 2015; 71(1). [7] Ning Chen, Xiao-ping Song, Yi Liu. Edge Detection Based on Biomimetic Pattern Recognition. TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(9). [8] Qu zhongshui. An Algorithm of Image Quality Assessment Based on Data Fitting of Image Histogram. TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(1). [9] Hong-an Li, Jie Zhang, Baosheng Kang. Image Deformation Based on Wavelet Filter and Control Curves. TELKOMNIKA Indonesian Journal of Electrical Engineering. 2014; 12(5).