Medium 9781601323248

Image Processing, Computer Vision, and Pattern Recognition

Views: 4017
Ratings: (0)

IPCV is an international conference that serves researchers, scholars, professionals, students, and academicians who are looking to both foster working relationships and gain access to the latest research results. It is being held jointly (same location and dates) with a number of other research conferences; namely, The 2014 World Congress in Computer Science, Computer Engineering, and Applied Computing (WORLDCOMP'14). The Congress is among the top five largest annual gathering of researchers in computer science, computer engineering and applied computing. We anticipate to have attendees from about 85 countries/territories. The 2014 Congress will be composed of research presentations, keynote lectures, invited presentations, tutorials, panel discussions, and poster presentations.

List price: $99.95

Your Price: $79.96

You Save: 20%

Remix
Remove
 

113 Slices

Format Buy Remix

Session - Computer Vision and Applications

PDF

 

An Algorithm for Mobile Vision-Based Localization of Skewed Nutrition Labels that Maximizes Specificity

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

3

An Algorithm for Mobile Vision-Based Localization of Skewed Nutrition Labels that Maximizes

Specificity

Vladimir Kulyukin

Christopher Blay

Department of Computer Science

Utah State University

Logan, UT, USA vladimir.kulyukin@usu.edu

Department of Computer Science

Utah State University

Logan, UT, USA chris.b.blay@gmail.com

Abstract—An algorithm is presented for mobile vision-based localization of skewed nutrition labels on grocery packages that maximizes specificity, i.e., the percentage of true negative matches out of all possible negative matches. The algorithm works on frames captured from the smartphone camera’s video stream and localizes nutrition labels skewed up to 35-40 degrees in either direction from the vertical axis of the captured frame. The algorithm uses three image processing methods: edge detection, line detection, and corner detection. The algorithm targets mediumto high-end mobile devices with single or quad-core ARM systems. Since cameras on these devices capture several frames per second, the algorithm is designed to minimize false positives rather than maximize true ones, because, at such frequent frame capture rates, it is far more important for the overall performance to minimize the processing time per frame. The algorithm is implemented on the Google Nexus 7 Android 4.3 smartphone.

 

Automated Hair Color Determination

PDF

10

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Automated Hair Color Determination

1

Daniel S. Rosen1, and Cambron Carter2

Director, Imaging Technology, GumGum Inc., Santa Monica, CA, US

2

Image Scientist, GumGum Inc., Santa Monica, CA, US

Abstract— The detection of human features utilizing computer vision techniques can provide significant information for exploitation of image content. Identification of human hair and its color is known to be of use for a variety of endeavors including targeting advertisements of hair care products. The daily volume of imagery which must be processed for advertising as well as the uncontrolled environment in which they are typically captured, negates the use of semi-automated techniques. A method of automated hair color determination which achieves high accuracy is presented.

relies on careful training of a location prior model. Finally, a technique has been developed using active shapes based on training a hair shape model [4]. While yielding good results, this technique is susceptible to error in cases of large shape variation, which may be caused by lighting effects or image geometry.

 

Automated Area Defect Inspection of Touch Panels Using Computer Vision

PDF

16

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Automated Area Defect Inspection of Touch Panels

Using Computer Vision

Hong-Dar Lin, Jen-Miao Li

Department of Industrial Engineering and Management, Chaoyang University of Technology,

Taichung, 41349, Taiwan

Abstract – Touch panels (TP) are widely used in various electronic products. It is a difficult inspection task when defects embedded in surfaces of TPs with structural textures.

A common surface defect type called area defects includes dirt, water marks, bubbles, and other defects with larger size.

Such defects have low contrast, brightness with gradual changes, irregular and non-directional shapes, and there may be both bright and dark flaws co-existing in a region.

Therefore, this study proposes an automated detection method to inspect the area defects on touch panels. The proposed method applies the Haar Wavelet transform with flat zone filtering operation to remove the structural textures of background through filtering an approximated sub-image of a decomposed image in wavelet domain. Then, the filtered image is transformed back to spatial domain. Finally, the restored image can be easily segmented to into three categories namely dark defects, bright defects, and background by using a simple statistical histogram method.

 

Traffic Control by Digital Imaging Cameras

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

23

Traffic Control by Digital Imaging Cameras

Rowa’a Jamal, Karmel Manaa, Maram Rabee’a, Loay Khalaf

Electrical Engineering Department, University of Jordan, Amman, Jordan

Abstract— Traffic Control is considered one of the fastest developing technologies in the world. One such method is control by traffic cameras. The first cameras installed for traffic monitoring were developed in the 1960s. This development led to the growth of multi-purpose traffic cameras in several countries across the world. The scope of this report is concentrated on producing a traffic control camera, which can be installed at crossroads with traffic lights. Algorithms for speed detection, and license plate recognition, are described and their performance is evaluated.

Keywords— Image processing, traffic camera, plate’s recognition, traffic.

1 Introduction

they have advantages over film cameras in speed monitoring.

However, film-based systems may provide superior image quality in the variety of lighting conditions encountered on roads. New film-based systems are still being used, but digital ones are providing greater proficiency, lower maintenance and are now more popular.

 

An Orange Sorting Technique based on Size and External Defects

PDF

30

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

An Orange Sorting Technique based on Size and

External Defects

Naeem Sattar, Sheikh Ziauddin, Sajida Kalsoom, Ahmad R. Shahid, Rafi Ullah, Amir H. Dar

Department of Computer Science

COMSATS Institute of Information Technology

Islamabad, Pakistan

Abstract—In this paper, a new automated orange sorting technique is presented. It sorts orange fruit based on size and external defects. We use image processing techniques for segmentation of oranges and then based on our algorithm categorize them according to the presence and type of defect. The defects considered are anthracnose, unripe and stem-end injury. Non-defected oranges are further categorized into large, medium or small based upon their size. We created an image dataset containing images of 189 oranges. We achieve high accuracy with success rates of 97.4% and 100% for defect-based and size-based sorting, respectively.

Index Terms—Precision agriculture, citrus grading, citrus sorting, computer vision in agriculture, image processing in agriculture

 

An Algorithm for In-Place Vision-Based Skewed 1D Barcode Scanning in the Cloud

PDF

36

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

An Algorithm for In-Place Vision-Based Skewed 1D

Barcode Scanning in the Cloud

Tanwir Zaman

Department of Computer Science

Utah State University

Logan, UT, USA tanwir.zaman@aggiemail.usu.edu

Vladimir Kulyukin

Department of Computer Science

Utah State University

Logan, UT, USA vladimir.kulyukin@usu.edu

Abstract—An algorithm is presented for in-place vision-based skewed 1D barcode scanning that requires no smartphone camera alignment. The algorithm is in-place in that it performs no rotation of input images to align localized barcodes for scanning. The algorithm is cloud-based, because image processing is done in the cloud. The algorithm is implemented in a distributed, cloud-based system. The system’s front end is a smartphone application that runs on Android 4.3 or higher. The system’s back end is currently deployed on a four node Linux cluster used for image recognition and data storage. The algorithm was evaluated on a set of 506 video recordings of common grocery products. The videos had a 1280 x 720 resolution, an average duration of 15 seconds, and were recorded on an Android Galaxy Nexus smartphone in a local supermarket.

 

Session - Imaging Science and Medical Applications

PDF

 

Sickle Anemia and Distorted Blood Cells Detection Using Hough Transform Based on Neural Network and Decision Tree

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

45

Sickle Anemia and Distorted Blood Cells Detection Using

Hough Transform Based on Neural Network and Decision

Tree

1

Hany A. Elsalamony1

Mathematics Department, Faculty of Science, Helwan University, Cairo, Egypt. h_salamony@yahoo.com

Abstract - Sickle-cell anemia is one of the most important common types of anemia disease. This paper presents proposed algorithm in two parts, one is the construct an algorithm can detecting and counting RBCs (benign or distorted) in a microscopic colored image; even if they are hidden or overlapped. Second part is checking and analysing the constructed data of RBCs by applying the most common important techniques in data mining; neural network and decision tree. The experimental results are demonstrated high accuracy, and success using these two models in predicting for all the benign or distorted cells. This algorithm has achieved the highest segmentation by about 99.98% of all input cells, which is contributed to improve the diagnosis of Sickle anemia.

 

Estimation of Resected Liver Regions Using a Tumor Domination Ratio

PDF

52

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Estimation of Resected Liver Regions

Using a Tumor Domination Ratio

Masanori Hariyama1 , Moe Okada1 , Mitsugi Shimoda2 , Keiichi Kubota2

1 Graduate School of Information Sciences, Tohoku University, Japan

2 Second Department of Surgery, Dokkyo Medical University

Abstract— This article presents an automatic approach to estimate optimal resected-liver regions for oncologic surgery planning. Usually, resected liver regions are determined by selecting cut points on the portal vessels on 3D simulation software. Since the liver has complex vessel structure, it is difficult for human to find optimal resected liver regions. To solve this problem, a tumor domination ratio is proposed to find all portal vessels related to tumors. The tumor domination ratio allows us to compute the ideal resected region, that is, all the perfusion territories related to the tumor. Moreover, some types of conditions such as the size of vessels are considered for practical surgical use.

 

Rebuilding IVUS Images From Raw Data of the RF Signal Exported by IVUS Equipment

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Rebuilding IVUS Images From Raw Data Of The RF

Signal Exported by IVUS Equipment

Marco Aurélio Granero¹,³, Marco Antônio Gutierrez², Eduardo Tavares Costa¹

¹ Department of Biomedical Engineering– DEB/FEEC/UNICAMP, Campinas, Brazil

² Division of Informatics/Heart Institute – HCFMUSP, São Paulo, Brazil

³ Federal Institute of Education, Science and Technology S. Paulo – IFSP, São Paulo, Bra

Abstract - The study of composition and classification of atherosclerotic plaque has been a very active research field, both in cardiology and image processing. Intravascular ultrasound (IVUS) is an effective tool, which can insights about the crosssection of blood vessels, with sufficient accuracy to allow an accurate assessment of CT slices. This enables information about blood vessel structures to be determined. During an IVUS medical examination, physicians subjectively adjust a set of parameters to improve the visualization of a Region Of Interest (ROI) and produce corresponding images in Digital Imaging and Communications (DICOM) format, for later analysis and study. DICOM is appropriate for storage, transportation and access, but limits subsequent changes to image parameters, such as contrast or brightness. This makes comparison across patient populations difficult and restricts image processing operations. This paper details an alternative to using DICOM, which is to rebuild IVUS images from raw radiofrequency signal (RF) data. The main advantage of this process is the independence of the acquisition parameters adjusted during the exam.

 

Image Segmentation Techniques Applied to Point Clouds of Dental Models with an Improvement in Semi-Automatic Teeth Segmentation

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

63

Image Segmentation Techniques Applied to Point Clouds of Dental Models with an Improvement in SemiAutomatic Teeth Segmentation

Tamayo-Quintero, J. D.1, Arboleda-Duque, S.1,2 and Gómez-Mendoza, J.B. 1

1

Department of Electric, Electronic and Computer Engineering, Universidad Nacional de Colombia, Manizales, Caldas,

Colombia

2

Department of Telecommunication Engineering, Universidad Catolica de Manizales, Manizales, Caldas, Colombia

Abstract - This paper presents an exploratory study on the application of a combination of different segmentation techniques to point clouds of dental models. The techniques are based in geometric primitives (e.g. RANSAC), region growing segmentation and graph theory (particularly the

"Min-Cut" algorithm), and were tested using dental 3D point clouds. Data were acquired using a Konica Minolta Vivid 9i laser range scanner.

Also, a semi-automatic segmentation methodology is presented. Results of teeth segmentation using testing data suggest that it is possible to automatically segment teeth from digital 3D models.

 

Automatic Mass Segmentation Method in Mammograms Based on Improved VFC Snake Model

PDF

70

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Automatic Mass Segmentation Method in Mammograms

Based on Improved VFC Snake Model

Xiangyu Lu, Yide Ma, Weiying Xie, and Tongqing Li

School of Information Sci. Eng, Lanzhou University, Lanzhou, China

Abstract - Mammography analysis is an efficient way for the early detection of breast cancer. In this paper, we present an integrated method for mass auto-segmentation in breast. First of all, the local threshold method, Rough Set theory and morphological filter are used to remove the label and enhance the mammogram. Secondly we apply the Hough Transformation algorithm on the pre-processed image and locate the lesion as an approximate parametric circle which would be used as the initial contour of Snake model followed by. Finally, the mass boundary is accurately segmented by the improved

VFC Snake. This approach is tested on DDSM and MIAS database and the detection rates are 91.47% and 85% respectively. The average area overlap ratio between our results and the ground truth on MIAS database reaches

 

Correction of Intensity Nonuniformity in Mammographic Images

PDF

76

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Correction of Intensity Nonuniformity in Mammographic Images

S. Yazdani1 , R. Yusof 2 , A. Karimian3, A. Hematian4

1,2

Centre for Artificial Intelligence & Robotics, University Technologi Malaysia, Kuala lumpur ,Malaysia

3

Department of Biomedical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran

4

Dept. of Advanced Informatics School (AIS), Universiti Teknologi Malaysia (UTM), Kuala lumpur ,Malaysia

Abstract- Breast cancer is one of the most prevalent cancers among women[1]. Mammography, as one of the primary studies, is used for diagnosis of breast disease. In addition MR images can depict most of the significant changes of breast during the time . For the first step of breast disease detection, the density measurement of the breast on MR Images may provide very useful information.

MR images have some instinctive limitations like the strongly dependence of contrast upon the way the image is acquired, intensity inhomogeneities (Bias Field), etc.

 

Session - Biometrics + Face Recognition, Expression Detection, Human Detection

PDF

 

Finger Vein Recognition in Row and Column Directions Using Two Dimensional Kernel Principal Component Analysis

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Finger Vein Recognition in Row and Column Directions

Using Two Dimensional Kernel Principal Component

Analysis

Sepehr Damavandinejadmonfared, Vijay Varadharajan

Advanced Cyber Security Research Centre

Dept. of Computing, Macquarie University

Sydney, Australia

Abstract - In this paper, a whole identification system is introduced for finger vein recognition. The proposed algorithm first maps the input data into kernel space, then; Two

Dimensional Principal Component Analysis is applied to extract the most valuable features from the mapped data.

Finally, Euclidian distance classifies the features and the final decision is made. Because of the natural shape of human fingers, the image matrixes are not square, which makes it possible to use kernel mappings in two different ways-along row or column directions. Although, some research has been done on the row and column direction through 2DPCA, our argument is how to map the input data in different directions and get a square matrix out of it to be analyzed by Two

 

Similarity Measures for Fingerprint Matching

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

89

Similarity Measures for Fingerprint Matching

Kareem Kamal A.Ghany1, Aboul Ella Hassanien2 and Gerald Schaefer3

1 Faculty of Computers and Information, Beni Suef University, Egypt

2 Faculty of Computers and Information, Cairo University, Egypt

3 Department of Computer Science, Loughborough University, U.K.

Abstract— In this paper, we investigate different distance metrics for measuring the similarity between fingerprint templates. In particular, we apply several of them during the matching phase of a fingerprint system, and evaluate the obtained results. Our experiments show the Dice coefficient to give the most convincing results with a matching score of

93%, a false rejection rate of 0.04 and a false acceptance rate of 0.006.

Keywords: Fingerprint matching, similarity measure, distance,

Dice coefficient.

1. Introduction

Concepts of similarity and distance are important in many applications. They are for example necessary to measure the similarity of different objects, and thus form an essential part in many pattern recognition applications that involve clustering, classification, recognition, or retrieval. With a large number of similarity measures having been introduced in the literature, selecting an appropriate one for a particular task is crucial, since the success of the related application may depend critically on this choice. Similarity measures vary depending on the data types used [1].

 

Edge Histogram Descriptor for Finger Vein Recognition

PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

Edge Histogram Descriptor for Finger Vein Recognition

1

Yu Lu1, Sook Yoon2, Daegyu Hwang1, and Dong Sun Park2

Division of Electronic and Information Engineering, Chonbuk National University, Jeonju, South Korea

2

Department of Multimedia Engineering, Mokpo National University, Jeonnam, South Korea

Abstract - Edge histogram descriptor (EHD) is an efficient

texture representation method originally proposed in MPEG7 to express the local edge distribution in an image. To efficiently utilize the edge and orientation features of rich veins located inside a finger, in this paper, we propose a finger vein recognition method using edge histogram descriptor. Different from the original usage that divides the image space into 4 h 4 sub-images, we investigate the relationship between finger vein recognition performance and partition style of input image. The optimal parameter is searched for final recognition. Additionally, the nearest neighbor classifier with Euclidean distance metric is employed for matching. Experimental results on an available finger vein database, MMCBNU_6000, show that the proposed method performs better than those using state-ofthe-art algorithms.

 

Load more


Details

Print Book
E-Books
Slices

Format name
PDF
Encrypted
No
Sku
B000000032169
Isbn
9781601323248
File size
35 MB
Printing
Allowed
Copying
Allowed
Read aloud
Allowed
Format name
PDF
Encrypted
No
Printing
Allowed
Copying
Allowed
Read aloud
Allowed
Sku
In metadata
Isbn
In metadata
File size
In metadata