Hindawi

Journal of Sensors
Volume 2017, Article ID 3964376, 7 pages
https://doi.org/10.1155/2017/3964376

Research Article
Aphid Identification and Counting Based on Smartphone and
Machine Vision

Suo Xuesong, Liu Zi, Sun Lei, Wang Jiao, and Zhao Yang
Mechanical and Electrical Engineering, Agricultural University of Hebei, Baoding, Hebei 071001, China

Correspondence should be addressed to Suo Xuesong; [email protected]

Received 27 April 2017; Revised 20 June 2017; Accepted 17 July 2017; Published 29 August 2017

Academic Editor: Oleg Sergiyenko

Copyright © 2017 Suo Xuesong et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Exact enumeration of aphids before the aphids outbreak can provide basis for precision spray. This paper designs counting software
that can be run on smartphones for real-time enumeration of aphids. As a first step of the method used in this paper, images of
the yellow sticky board that is aiming to catch insects are segmented from complex background by using GrabCut method; then
the images will be normalized by perspective transformation method. The second step is the pretreatment on the images; images
of aphids will be segmented by using OSTU threshold method after the effect of random illumination is eliminated by single image
difference method. The last step of the method is aphids’ recognition and counting according to area feature of aphids after extracting
contours of aphids by contour detection method. At last, the result of the experiment proves that the effect of random illumination
can be effectively eliminated by using single image difference method. The counting accuracy in greenhouse is above 95%, while
it can reach 92.5% outside. Thus, it can be seen that the counting software designed in this paper can realize exact enumeration of
aphids under complicated illumination which can be used widely. The design method proposed in this paper can provide basis for
precision spray according to its effective detection insects.

1. Introduction Yue-hua et al. [7] extracted the aphids by using region

growing algorithm and OSTU threshold algorithm in the nat-
As the main representative of crop pests affecting wheat field ural environment, and the recognition rate of vector machine
and orchard fruit output [1], aphids with a huge number are training images can reach 90.7% by using 𝑘-means; Bai-jing
widely distributed. By hanging the yellow sticky board on et al. [8] extracted the aphids on the use of the method of
plants, people can take count of the aphids on the board so
G component threshold (each pixel in the aphids and leaves)
that they can determine if pesticide is needed. However, this
and carried out counting research on the use of connected
method is imprecise.
area markers. Jingxiao [9] transmitted the image of the pest
Compared with manual counting, machine vision can
calculate the number of pests objectively, accurately, and to the PC side for HSV color model segmentation and then
effectively through analyzing the image information. To solve the pest is identified by BP neural network. Vincent et al. [10]
these problems, a lot of research has been carried out. proposed a target detector that uses the analytical decision
Cassani et al. approached automated EEG-based Alzheim- support system to make the detector adapt to changes in the
er’s disease diagnosis using relevance vector machines [2], plant to identify aphids on the plant; Tirelli et al. studied
Chakraborty and Newton studied climate change, plant dis- the remote monitoring system to prevent pest outbreaks and
eases, and food security [3], Pal and Foody used evaluation of trigger an alarm when the number of pests is too high [11];
SVM, RVM, and SMLR for accurate image classification with Ukatsu achieved the remote counting function of rice margin
limited ground data [4], Patel et al. used improved multiple bugs through using the wireless network technology.
features based algorithm for fruit detection [5], Luo et al. These methods reflect the following problems: they will
develop an aphid damage hyperspectral index for detecting be influenced by illumination and the threshold is not
aphid (Hemiptera: Aphididae) damage levels in winter wheat unique in threshold extraction method. Extracting aphids by
and so on [6]. PC remoting method costs too much network traffic with
2 Journal of Sensors

poor timeliness. Therefore, to obtain images in the complex 𝜃 makes the boundary energy more and more little by
environment and high impact of the scene, pests can be iterative method; after iteration, the parameters of the Gauss
identified and counted without relying on the feedback of the mixture model of target and background are higher.
PC side.
𝜃 = {𝜋 (𝛼, 𝑝) , 𝜇 (𝛼, 𝑝) , ∑ (𝛼, 𝑝) , 𝛼 = 0, 1, 𝑘
2. Materials and Methods (4)
= 1 ⋅ ⋅ ⋅ 𝑘} .
The acquisition of experimental images was obtained at the
West Campus of Agricultural University of Hebei; image 2.2. Correction of Yellow Sticky Board. Because the shooting
acquisition equipment is Samsung Galaxy 4 phone that has angle is not fixed, the positions of yellow sticky board which
a Qualcomm eight-core processor with Android 2.3 system, has been extracted are not fixed. In order to improve the
API 25. It is a common type with satisfactory parameters. accuracy of aphid identification and counting, the yellow
The chapter contains two parts of picture distillation; in sticky board must be corrected to the same size and the same
the first part, the GrabCut algorithm [12] is used to extract position when a target image has been segmented; getting the
yellow sticky board image from complex background and elevation view, the aphids can be counted accurately.
correct it to the same size and position by perspective Transformation from distortion to elevation view can
correction method in order to extract aphid image [13]; in be completed by using perspective correction method [14]
the second part, the method of weighted average gray scale which uses the four vertexes of distortion image and elevation
is used to make the yellow sticky board image preprocessing, view as the correction basis. The top left vertex (𝑥, 𝑦) is the key
single image difference method is used to eliminate the of correction. Firstly, the corrected image matrix [𝑥2 𝑦2 𝑧2 ]
influence of illumination, and OSTU threshold segmentation is obtained by using the distorted image matrix [𝑥1 𝑦1 𝑧1 ];
and morphological erosion method is used to extract yellow you can get the top left vertex (𝑥, 𝑦) by using the relationship
sticky board image, laying the groundwork for the subsequent between [𝑥1 𝑦1 𝑧1 ]. Secondly, according to top left vertex,
counting program. the length and width of the rectangle can be used to obtain
the information of four vertices.
2.1. Extraction the of Yellow Sticky Board Image. GrabCut 𝑎11 𝑎12 𝑎13
is an improved algorithm based on GraphCut. GrabCut is [ ]
𝐷 = [𝑥2 𝑦2 𝑧2 ] = [𝑥2 𝑦2 𝑧2 ] [𝑎21 𝑎22 𝑎23 ] ,
an algorithm that uses the RGB three-channel hybrid Gauss
model for color segmentation; segmentation can be com- [𝑎31 𝑎32 𝑎33 ]
pleted by using region pixel set which contains background. 𝑥 𝑎 𝑥 + 𝑎21 𝑦1 + 𝑎31 (5)
The algorithm belongs to iterative segmentation, and the 𝑥 = 2 = 11 1 ,
𝑧2 𝑎13 𝑥1 + 𝑎23 𝑦1 + 𝑎33
segmentation accuracy is higher. Gibbs energy function is a
specific vocabulary that means a formula that can calculate 𝑦2 𝑎12 𝑥1 + 𝑎22 𝑦1 + 𝑎32
𝑥= = .
the weights and cost. 𝑧2 𝑎13 𝑥1 + 𝑎23 𝑦1 + 𝑎33

𝐸 (𝛼, 𝑝, 𝜃, 𝑧) = 𝑈 (𝛽, 𝑝, 𝜃, 𝑧) + 𝑉 (𝛼, 𝑧) . (1) Figure 1 is the elevation view of yellow sticky board by
using GrabCut and perspective correction method. We can
see that the picture has a clear boundary and regular shape
𝑈 is a method when the target pixels are classified as which will make the aphids extracted easily.
foreground or background models, which are obtained by
matrix 𝐷 iteration of classifying and rewriting pixel: 2.3. The Image of the Yellow Sticky Board Preprocessing. The
color image change into gray image can reduce amount of
𝑈 (𝛼, 𝑝, 𝜃, 𝑧) = ∑ 𝐷 (𝛼𝑛 , 𝑝𝑛 , 𝜃, 𝑧𝑛 ) , computation in the image processing process and reduce the
𝑛 complexity of the calculation [15, 16], the gray scale value
1 obtained by the weighted gray scale method is more in line
𝐷 (𝛼𝑛 , 𝑝𝑛 , 𝜃, 𝑧𝑛 ) = log 𝜋 (𝛼𝑛 , 𝑝𝑛 ) + log det ∑ (𝛼𝑛 , 𝑝𝑛 ) with the perception of the corresponding gray scale of each
2 (2) pixel, and the brightness is moderate; it is conducive to the
1 −1 next aphid image extraction.
+ [𝑧 − 𝜇 (𝛼𝑛 , 𝑝𝑛 ) ]
2 𝑛
−1 −1 2.4. Aphids Extraction
⋅ ∑ (𝛼𝑛 , 𝑝𝑛 ) [𝑧𝑛 − 𝜇 (𝛼𝑛 , 𝑝𝑛 )] .
2.4.1. Image Difference. Due to the influence of light, even
𝑉 is the boundary energy smoothing term, when the if more than two thresholds segmentations are proposed;
boundary pixels are discontinuous; it will do some boundary each of the pictures taken in the natural environment has
pixel classification processing: an uncertain gray value, so the threshold is uncertain too.
Aphid color is relatively single, so the impact of light is
󵄩 󵄩2 mainly reflected in the background; an image can be seen to
𝑉 (𝛼, 𝑧) = 𝛾 ∑ [𝛼𝑥 , 𝛼𝑦 ] exp − 𝛽 󵄩󵄩󵄩󵄩𝑧𝑥 − 𝑧𝑦 󵄩󵄩󵄩󵄩 . (3) consist of foreground and background; when the background
(𝑥,𝑦)∈𝑐 is removed, the influence of light is removed. Regarding the
Journal of Sensors 3

Figure 1: Perspective correction. Figure 2: Image difference method for extracting aphids.

process of image difference, firstly the original gray image

will be taken for fuzzification as SRC and then subtract the
fuzzy background DST. The difference that results, DIFF, is
the aphids image.
SRC − DST = DIFF. (6)

2.4.2. Fuzzy Background. The gray scale image after blurring

is compared with the gray scale distribution in the original Figure 3: A digital mark of a binary image.
gray scale graph, and if the local gray scale distribution is
equal, it can be used as the background image. Compared
with the fuzzy effect of the three filters, the Gaussian weight 2.5. Aphids Count
filter has the smallest noise, which is consistent with the
original gray scale distribution. When the kernel value is (51, 2.5.1. Feature Extraction. In order to count accurately, it is
51), the background blurring is the best. necessary to identify whether there are aphids. With access
to relevant information we can see that the aphid’s length is
2.4.3. Threshold Selection. Image difference is only gray between 1.5 mm and 4.9 mm. Most aphids are 2 mm in length,
image, and the gray image is not the result of image seg- and there are differences in the lengths of other insects in the
mentation and needs the threshold segmentation [17]. A orchard; for example, the apple leaf mites are about 0.3 mm.
large number of aphids are missing counted by using binary So distinguish between aphids and nonaphids according to
threshold method, while there are too many contour break area characteristics. First, all the contours in the binary image
points by using self-adaption threshold method. So this paper are identified as digital marks, and the sum of the pixels in
uses the OSTU threshold segmentation method which has a each connection domain is calculated to find the relationship
clear and integrated contour as the binarization method. between the actual length of the aphids and the pixel values,
as shown in Figure 3.
According to Table 1, based on the relationship between
2.4.4. Morphological Operation. In the natural environment, the pixel values in each connected domain and the actual
the aphids on the yellow sticky board have a high base and lengths of aphids, we can draw a conclusion that when the
a high density, and some of the aphids’ adhesion does not pixel value is less than 20, the insect’s body length increases by
affect the overall counting result and requires some corrosion 0.1 mm and the pixel value is increased by 1 px∼2 px; when the
of the partially adhered aphids to reduce the error. This pixel value is higher than 20, the insect’s body length increases
paper uses the elliptical nucleus with a nuclear value of (4, 4) by 0.1 mm and the pixel value is increased by 3 px∼4 px. In
for corrosion operation. This method has a good effect on this paper, the aphids involved are mostly 1.5 mm to 2.8 mm,
elimination of adhesions. so when the aphid area characteristics range from 15 px to
𝐸 (𝑥) = {𝑎 | 𝐵𝑎 ∈ 𝑋} = 𝑋Θ𝐵. (7) 48 px, nonaphids range from 1 px to 10 px is the most suitable.
According to the range of aphid area characteristics derived
Figure 2 shows the use of image difference algorithm to from the above study to identify aphid images on the yellow
obtain the aphid images; aphid images are clear and complete. sticky board [18], in the obtained result graph, the aphid
4 Journal of Sensors

Table 1: Part of the connection domain number, pixel values, and Main
the relationship between the lengths of pests. interface

The pixel value within the contour (contour

Pest category
number, pixel value, and aphid length/mm) Retrieve the
Call the local local album
(29, 41, 2.6) (31, 28, 2.2) (32, 35, 2.5) camera image
(34, 21, 2) (35, 19, 1.9) (36, 19, 1.9) Image
acquisition
(37, 20, 1.9) (38, 21, 1.9) (39, 27, 2.2) module
(43, 32, 2.2) (44, 29, 2.2) (45, 19, 1.8)
Aphid Image
(47, 20, 1.9) (48, 20, 1.9) (49, 42, 2.6)
compression
(50, 18, 1.9) (51, 40, 2.6) (53, 35, 2.5)
(56, 25, 2.1) (58, 19, 1.8) (60, 23, 2.1)
(61, 48, 2.8) (63, 32, 2.4) (64, 19, 1.9) Extraction of
(30, 2, 0.2) (41, 4, 0.4) (52, 6, 0.6) yellow sticky board
Nonaphids (42, 9, 0.9) (54, 8, 0.9) (57, 10, 1)
(59, 9, 0.9) (65, 10, 1) Correction of Image
yellow sticky processing
board module
image is marked with white and nonaphid images are not
marked with white, consistent with the human eye.
Aphid extraction
2.5.2. Aphids Counting Method. Using findContours() con-
tour extraction method to find all the contours of the image.
Firstly, the outline which can be detected in the image will
be converted to vector form to iterate, and then the aphids Identification and Image counting
counting of aphids module
section will be screened out by using “for-cycle” according to
their area features [19].
Figure 4: Software flow chart.
2.6. Design and Implementation Based on
the Android Platform
loaded into Imageview by using decodeFile method. At last
2.6.1. Software Overall Flow Chart. In order to meet the needs the image information will return and be displayed on the
of the public for the convenience of equipment, Android next interface.
system as a mobile platform mainstream system [20] in the
mobile side of the software development plays a guiding (ii) Image Processing Module. In the image processing mod-
role; based on the Android platform aphid count APP will ule, we set “yellow sticky board segmentation” “yellow
greatly facilitate the majority of farmers and agricultural sticky board correction” “aphids identification” to extract the
science and technology workers. The aphid counting soft- aphids.
ware in this paper uses the Java programming language Using GrabCut algorithm to extract yellow sticky board,
in the Android studio platform with OpenCV computer first create a Rect object to create a rectangular box to
vision library development for Android 4.0.3 and above. The select the possible prospects in the image and use the
software implements the counting of aphids; the whole is canvas.drawRect () method to draw the rectangle and pass
divided into three modules: image acquisition module, image into the mask matrix. Create a mask object as the display
processing module, and image counting module; as shown position of the segmented yellow sticky board image, which
in Figure 4, now we design and implement the part of the is the image of the CvType.CV 8UC1 channel that needs to
content of the software. be of the same size as the original image. Black is used as the
color of scallar which means the background of output image
(i) Image Acquisition Module. The software obtains the picture is black as well. Finally the segmentation can be completed by
in two ways: invoking the local camera to shoot in real calling the arguments.
time or retrieving the existing photo in the local album. Use the perspectiveTransform () method to correct the
After the photo is taken to ensure the quality of the picture, yellow sticky board. Use the “for-cycle” to traverse the four
the appropriate image compression can not only reduce the edge vertexes of the two intersecting lines in the segmented
computational complexity of the image processing process image by using GrabCut. Then compare the coordinates of
but also avoid the overflow caused by the image being the four vertexes with the midpoint of the yellow sticky board.
too large to cause the program to crash. First create an For example, in the coordinates (𝑥, 𝑦) of the horizontal axis,
intent object and set the action of the object as the string the value of the vertical axis is less than the corresponding
of MediaStore, ACTION IMAGE CAPTURE. Then use the value of the midpoint; then the point is the top left vertex.
string of startActivityForResult to start Intent program. After According to the upper left, upper right, lower left, and lower
setting the length and width, the image information will be right of the order of the four vertex coordinates in the variable
Journal of Sensors 5

(a) In greenhouse (b) Orchard

Figure 5: Practical application of aphid counting software.

called startM, the perspective transformation method can be 140

Equation y= a+b∗x
adj. R2 0.99504
corrected. Value Standard error
Use single image difference method to extract aphids. 130 Manual counts Intercept −2.29425 1.83911
number
Through the Imgproc class cvtColor, color image will be Manual counts Slope 1.04846 0.02468
120
converted to the output image grayMat gray image. Set number
Manual counts number

the parameters of nuclear value Size (𝑤, ℎ) as Size (51, 51); 110
then use the Imgproc.GaussianBlur method to complete
100
the process of fuzzy background. Use Core.absdiff() for
differential operation. Use the Imgproc.THRESH OTSU for 90
threshold operation after difference background image. The
differential images need to undergo morphological corrosion 80
to remove the effects of the aphids. Set the morphological 70
action as MORPH ERODE and set the shape of structure
element kernel1 as MORPH ELLIPSE with a new Size (5, 5) 60
by using getStructuringElement(). Elimination of adhesions 50
will be completed.
50 60 70 80 90 100 110 120 130
(iii) Image Counting Module. The image counting module APP counts number
filters the contours based on the total pixel range in each
Greenhouse
connected domain and stores the count on the counter. The
Orchard
output value is the number of aphids. Linear fit of greenhouse
The counting process can be seen as a traversal of the
“funnel”; then we should define two containers: the first con- Figure 6: The comparison of accuracy between the manual count
tainer is stored in all the contour data, whether or not meeting and the APP count.
the characteristics of the requirements, and the second con-
tainer is stored after the screening of the contour information.
Traverse the contour type to Imgproc.RETR CCOMP and the counting results of the aphids in the greenhouse, and
output it as Imgproc.CHAIN APPROX SIMPLE. The “for- Figure 5(b) shows the counting results of the aphids in the
cycle” is the key of selecting. Set 15 px to 18 px as the shape of orchard.
aphids image in the connected domain and set scallar (255, Based on the least squares method, a linear regression
255, 255) to show aphids as white. analysis has been carried out by taking 10 sets of data from
Table 2 in the greenhouse to get the fitting curve. 𝑥-axis is
3. Results and Discussions the APP counts number and 𝑦-axis is the manual counts
number. As shown in Figure 6, the relationship of the fitting
Using the aphid counting software to count the images in between the mobile APP count and the manual count is
greenhouse and the images in orchards, Figure 5(a) shows 𝑦 = 1.04846𝑥 − 2.29425; the fit degree 𝑅2 factor is up to
6 Journal of Sensors

Table 2: Error analysis of manual count and APP count.

Image source Image number Number of manual counts/month APP counts number/month Relative error/% Accuracy/%
(1) 52 52 0 100
(2) 80 79 1.25 98.75
(3) 62 61 1.61 98.38
(4) 84 82 2.38 97.61
(5) 86 87 1.16 98.83
Greenhouse
(6) 55 53 3.7 96.22
(7) 54 54 0 100
(8) 69 67 2.9 97.01
(9) 70 70 0 100
(10) 121 116 4.13 95.86
(11) 71 66 7.04 92.95
Orchard
(12) 138 128 7.24 92.75

0.99504, indicating that the aphid recognition and mobile and convenient. The method proposed in this paper can
APP counting can be used to achieve functions of recognition extract and correct the image of yellow sticky board, eliminate
and counting. Adding two sets of orchard data from Table 2 influence of illumination on the image, and count aphids
into the coordinate system, we can find that the two points accurately.
are in the vicinity of the curve, indicating that the method is Due to the limitations of the season and the study site, the
basically effective. Because the orchard environment is more aphid counting software needs to design a memory storage
complex, the accuracy rate of the method is slightly worse module to achieve the same number of yellow sticky insect
than the greenhouse; as shown in Table 2, the greenhouse boards to make trend prediction function.
accuracy rate is up to 95% and orchard accuracy rate is up
to 92%. In general, the aphids identification and counting Conflicts of Interest
method is highly reliable.
The authors declare that they have no conflicts of interest.
4. Conclusion
Acknowledgments
In the past, aphid identification focuses on the separation of
a single kind of aphids, few on different kinds of pests. The This work is supported by the new facilities of agricultural
count number is less, and the aphids’ antennae, foot, and machinery research and development (17227206D).
wings need to be extracted precisely, because the operation
complexity is high; most of the application software that References
implements the aphid recognition and counting function is
concentrated on the PC side. The traditional counting APP [1] X. Zhan-li, L. Xiao-xia, Z. Qing-wen et al., “Characteristics of
collects the photos through the Internet to transfer data by aphid age identification of wheat long pipe,” Acta Entomologica
the PC processing, and then the results return back to the Sinica, vol. 01, pp. 81–87, 2014.
mobile App, the process Overreliance on the PC side and [2] R. Cassani, T. H. Falk, F. J. Fraga, P. A. Kanda, and R. Anghinah,
the Internet; its real time performance is poor. Collection of “Towards automated EEG-Based Alzheimer’s disease diagnosis
aphids requires special equipment; this professional equip- using relevance vector machines,” in Proceedings of the 5th
ment is not convenient and also very expensive. The device ISSNIP - IEEE Biosignals and Biorobotics Conference (BRC ’14),
that collects the aphids is not universal. pp. 1–6, IEEE, May 2014.
This paper uses yellow sticky board in the natural [3] S. Chakraborty and A. C. Newton, “Climate change, plant
diseases and food security: an overview,” Plant Pathology, vol.
environment to attracts aphids and some other pests, and
60, no. 1, pp. 2–14, 2011.
collects images in the real growth environment of plants.
[4] M. Pal and G. M. Foody, “Evaluation of SVM, RVM and SMLR
Aphid counting software overcomes the complex natural
for accurate image classification with limited ground data,” IEEE
environment and the impact of light caused by image pro- Journal of Selected Topics in Applied Earth Observations and
cessing difficulties; it does not rely on the PC side and the Remote Sensing, vol. 5, no. 5, pp. 1344–1355, 2012.
network feedback; it can quickly identify and count aphids, [5] H. N. Patel, R. K. Jain, and M. V. Joshi, “Fruit detection using
which are the representative of the pest group; it has very improved multiple features based algorithm,” International
practical value. The accuracy of the aphid counting software Journal of Computer Applications, vol. 13, no. 2, pp. 1–5, 2011.
is compared with the result of manual counting; the accuracy [6] J. Luo, D. Wang, Y. Dong, W. Huang, and J. Wang, “Developing
rate of the aphid counting software is more than 95% in an aphid damage hyperspectral index for detecting aphid
the greenhouse environment; the accuracy rate of the aphid (Hemiptera: Aphididae) damage levels in winter wheat,” in
counting software is more than 92%. The operation is smooth Proceedings of the 2011 IEEE International Geoscience and
Journal of Sensors 7

Remote Sensing Symposium (IGARSS ’11), pp. 1744–1747, IEEE,

July 2011.
[7] C. Yue-hua, H. U. Xiao-guang, and Z. Chang-li, “Study on wheat
pest segmentation algorithm based on machine vision,” Journal
of Agricultural Engineering, vol. 12, pp. 187–191, 2007.
[8] Q. Bai-jing, W. Tian-bo, L. Juan-juan et al., “Image recognition
and counting method of cucumber aphids,” Journal of Agricul-
tural Mechanization, vol. 08, pp. 151–155, 2010.
[9] R. Jingxiao, Image-based pest automatic counting and recognition
system, Zhejiang University, 2014.
[10] M. Vincent et al., “Towards a Video Camera Network for Early
Pest Detection in Greenhouses,” 2008, http://www-sop.inria.fr/
pulsar/projects/bioserre/file/endure.pdf.
[11] P. Tirelli, N. A. Borghese, F. Pedersini, G. Galassi, and R.
Oberti, “Automatic monitoring of pest insects traps by Zigbee-
based wireless networking of image sensors,” in Proceedings of
the 2011 IEEE International Instrumentation and Measurement
Technology Conference (I2MTC ’11), pp. 1–5, IEEE, May 2011.
[12] J. Wang, L. Ji, A. Liang, and D. Yuan, “The identification of but-
terfly families using content-based image retrieval,” Biosystems
Engineering, vol. 111, no. 1, pp. 24–32, 2012.
[13] H. Ran, Image Segmentation Algorithm Based on Visual Atten-
tion Mechanism and Its Application, Beijing Jiaotong University,
2016.
[14] D. Qin, W. Yanjie, and H. Han Guangliang, “Perspective cor-
rection based on improved Hough transform and perspective
transformation,” Journal of Liquid Crystals and Display, vol. 04,
pp. 552–556, 2012.
[15] J. Faria, T. Martins, M. Ferreira, and C. Santos, “A computer
vision system for color grading wood boards using Fuzzy Logic,”
in Proceedings of the 2008 IEEE International Symposium on
Industrial Electronics (ISIE ’08), pp. 1082–1087, July 2008.
[16] Z. Jian-wei, W. Yong-mo, and S. Zu-ruo, “Study on automatic
counting of aphids in wheat field,” Chinese Journal of Agricul-
tural Engineering, vol. 09, pp. 159–162, 2006.
[17] W. Zhi-bin, W. Kai-yi, Z. Shui-fa, and M. Cui-xia, “Class
counting algorithm of whitefly based on K-means clustering
and elliptic fitting method,” Journal of Agricultural Engineering,
vol. 01, pp. 105–112, 2014.
[18] B. Kui, J. Chao, L. Lei, and W. Cheng, “Non-nondestructive
measurement of wheat morphological characteristics based
on morphological image processing,” Journal of Agricultural
Engineering, vol. 12, pp. 212–216, 2010.
[19] N. Gonçalves, V. Carvalho, M. Belsley, R. M. Vasconcelos, F.
O. Soares, and J. Machado, “Yarn features extraction using
image processing and computer vision—a study with cotton and
polyester yarns,” Measurement, vol. 68, pp. 1–15, 2015.
[20] W. Huaxu, “Development and design of software framework for
android platform image processing software,” Software, vol. 02,
pp. 46-47, 2014.
 International Journal of

Rotating
Machinery

International Journal of
The Scientific
(QJLQHHULQJ Distributed
Journal of
Journal of

Hindawi Publishing Corporation

World Journal
Hindawi Publishing Corporation Hindawi Publishing Corporation
Sensors
Hindawi Publishing Corporation
Sensor Networks
Hindawi Publishing Corporation
http://www.hindawi.com Volume 201 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of

Control Science
and Engineering

Advances in
Civil Engineering
Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

https://www.hindawi.com

Journal of
Journal of Electrical and Computer
Robotics
Hindawi Publishing Corporation
Engineering
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

VLSI Design
Advances in
OptoElectronics
,QWHUQDWLRQDO-RXUQDORI

International Journal of
Modelling &
Simulation
$HURVSDFH
Hindawi Publishing Corporation Volume 2014
Navigation and
Observation
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
in Engineering
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014
(QJLQHHULQJ
+LQGDZL3XEOLVKLQJ&RUSRUDWLRQ
KWWSZZZKLQGDZLFRP 9ROXPH
Hindawi Publishing Corporation
http://www.hindawi.com
http://www.hindawi.com Volume 201-

International Journal of
International Journal of Antennas and Active and Passive Advances in
Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration
Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014