2-D Image Matching Based on Global Harmony Search Optimization

2012 ◽  
Vol 239-240 ◽  
pp. 1133-1137 ◽  
Author(s):  
Li Ming Guan ◽  
Qian Kai Yang ◽  
Jian Lin ◽  
Yi Fan Wu
Keyword(s):  
Hausdorff Distance ◽  
Image Matching ◽  
Fitness Function ◽  
Harmony Search ◽  
Two Dimensional ◽  
Edge Detector ◽  
Matching Method ◽  
Two Dimensional Image ◽  
Search Optimization ◽  
Canny Edge

A two-dimensional image matching method based on the improved Hausdorff distance and global harmony search optimization is presented in this paper. First, edges are extracted form the original images by the Canny edge detector. Then, a fitness function based on the improved Hausdorff distance is constructed. Finally the global harmony search optimization is adopted to optimize the fitness function. Experiments show that the proposed method is able to locate the object of interest globally and efficiently.

Hybrid Inception Recurrent Residual Convolutional Neural Network (HIRResCNN) with Harmony Search Optimization (HSO) for Early Breast Cancer Detection System

2021 ◽  
Vol 19 (7) ◽  
pp. 01-24
Author(s):  
K. Sangeetha ◽  
S. Prakash
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Convolutional Neural Network ◽  
Detection System ◽  
Harmony Search ◽  
Principal Component ◽  
Edge Detector ◽  
Canny Edge Detector ◽  
Search Optimization ◽  
Canny Edge

For women, most common cause of death is Breast tumour and in worldwide, it is the second leading reason for cancer deaths. Due the requirement of breast cancer’s early detection and false diagnosis impact on patients, made researchers to investigate Deep Learning (DL) techniques for mammograms. There are four stages in this proposed HIRResCNN framework, namely, Pre-processing, reduction of dimensionality, segmentation and classification. From images, noises are removed using two filtering algorithms called Median and mean filtering in pre-processing stage. Then canny edge detector is used for detecting edges. Gaussian filtering is used in canny edge detector to smoothen the images. In the next dimensionality reduction stage, attributes are correlated using Principal Component Analysis (PCA) inclusive of related features. So, this huge dataset is minimized and only few variables are used for expressing it. In order to detect the breast cancer accurately, foreground and background subtraction is done in the third stage called segmentation stage. At last, for detecting and classifying breast cancer, a Hybrid Inception Recurrent Residual Convolutional Neural Network (HIRResCNN) is introduced, which integrates Harmony Search Optimization (HSO) to tune bias and weight parameters and classification accuracy is enhanced using HIRResCNN-HSO model. Strength of Recurrent Convolutional Neural Network (RCNN), Residual Network (ResNet) and Inception Network (Inception-v4), are combined in a powerful Deep Convolutional Neural Network (DCNN) model called HIRResCNN. using Mammographic Image Analysis Society (MIAS) dataset, various experiments are conduced and results are compared with other available techniques. Around 92.6% accuracy rate is produced using this proposed HIRResCNN classifier in finding breast cancer.

Image Matching Based on Bacterial Foraging Algorithm

2011 ◽  
Vol 301-303 ◽  
pp. 859-863
Author(s):  
Hong Peng Tian
Keyword(s):  
Wavelet Transform ◽  
Swarm Intelligence ◽  
Image Matching ◽  
Fitness Function ◽  
Optimal Solution ◽  
Experimental Results ◽  
Matching Method ◽  
Bacterial Foraging ◽  
Template Image ◽  
Bacterial Foraging Algorithm

To increase the speed of image matching, this paper combines Bacterial Foraging Algorithm (BFA) of swarm intelligence with wavelet transform, and presents a fast matching method. The method regards the problem of image matching as a search for the optimal solution. To provide artificial bacterial swarm algorithm with an appropriate fitness function, the Normalized Product correlation (NPROD) is employed to measure the similarity between the template image and the searching image. Then the best coarse matching position is gradually approaching by chemotaxis, elimination and dispersal, and reproduction behaviors of artificial bacterial. Finally, the best matching position is found out according to the coarse matching position. Experimental results show that the proposed method is fast and efficient.

Image matching method by using Hausdorff distance based on vector length

2013 ◽  
Vol 32 (11) ◽  
pp. 3161-3163
Author(s):  
Wei ZHANG ◽  
He-ping CHEN ◽  
Ling-xian YANG
Keyword(s):  
Hausdorff Distance ◽  
Image Matching ◽  
Matching Method ◽  
Vector Length

Aesthetic evolutionary algorithm for fractal-based user-centered jewelry design

2007 ◽  
Vol 22 (1) ◽  
pp. 19-39 ◽  
Author(s):  
Somlak Wannarumon ◽  
Erik L.J. Bohez ◽  
Kittinan Annanon
Keyword(s):  
Evolutionary Algorithm ◽  
Iterated Function System ◽  
Measure Theory ◽  
Fitness Function ◽  
Two Dimensional ◽  
User Centered Design ◽  
Two Dimensional Image ◽  
Art Forms ◽  
The Aesthetic ◽  
User Centered

AbstractThis paper proposes an aesthetic-driven evolutionary algorithm for user-centered design. The evolutionary algorithm is based on a genetic algorithm (GA). It is developed to work as an art form generator that enhances user's productivity and creativity through reproduction, evaluation, and selection. Users can input their preferences and guide the generating direction to the system. A two-step fitness function is developed to evaluate morphology and aesthetics of the generated art forms. Fractals created by an iterated function system are used for representing art forms in our process. Algorithmic aesthetics are developed based on the aesthetic measure theory, surveys of human preferences, and popular long-lasting symbols. The algorithmic aesthetics is used for evaluating aesthetics of art forms together with subjective nonquantifiable aspects, and placed in the fitness function. The GA basically creates two-dimensional art forms. However, any two-dimensional image can be included through the property of a condensation set of fractals. The proposed GA can increase design productivity by about 80%. Examples of jewelry designs and physical prototypes created by the proposed system are included.

Short Hand Recognition using Canny Edge Detector

2017 ◽  
Vol 7 (5) ◽  
pp. 902-907
Author(s):  
Pramod Kumar S ◽  
◽  
Narendra T.V ◽  
Vinay N.A ◽  
◽  
...  
Keyword(s):  
Edge Detector ◽  
Canny Edge Detector ◽  
Canny Edge ◽  
Short Hand ◽  
Hand Recognition

PERBANDINGAN METODE HOMOMORPHIC FILTERING DAN METODE CONTRAST STRECHING UNTUK PERBAIKAN KUALITAS CITRA UNDERWATER

2018 ◽  
Vol 2 (1) ◽  
Author(s):  
Adi Mora Lubis ◽  
Nelly Astuti Hasibuan ◽  
Imam Saputra
Keyword(s):  
Digital Image ◽  
Water Environment ◽  
Repair Process ◽  
Image Process ◽  
Two Dimensional ◽  
Two Dimensional Image ◽  
Underwater Image ◽  
Homomorphic Filtering ◽  
Contrast Stretching

Digital imagery is a two-dimensional image process through a digital computer that is used to manipulate and modify images in various ways. Photos are examples of two-dimensional images that can be processed easily. Each photo in the form of a digital image can be processed through a specific software. In the water environment, the light factor greatly influences the results of the quality of the image obtained. With the deepening of underwater shooting, the results obtained will be the darker the quality of the underwater image. . uneven lighting and bluish tones. One of the factors that influence the recognition results in pattern recognition is the quality of the image that is inputted. The image acquired from the source does not always have good quality. The process of repairing digital images that experience interference in lighting. The lighting repair process uses homomorphic filtering and uses contrast striching and will compare the quality of both methods and test to prove the results of image quality between homomorphic filtering and contrast streching. Until later the results of both methods can be seen which is better. homomorphic filtering and contrast stretching can produce image improvements with pretty good performance.Keywords: Digital Image, Underwater Image, Homomorphic Filtering, Contrast Streching, Matlab R2010a

IMPLEMENTASI METODE RETINEX UNTUK MENINGKATKAN KUALITAS CITRA UNDERWATER

2018 ◽  
Vol 2 (1) ◽  
Author(s):  
Bainun Harahap
Keyword(s):  
Water Environment ◽  
Repair Process ◽  
Image Process ◽  
Two Dimensional ◽  
Two Dimensional Image ◽  
Underwater Image ◽  
Object Search ◽  
Image Improvement ◽  
Underwater Object

Digital imagery is a two-dimensional image process through a digital computer that is used to manipulate and modify images in various ways. Photos are examples of two-dimensional images that can be processed easily. Each photo in the form of a digital image can be processed through certain software devices. In the water environment, light factors greatly influence the results of image quality obtained. With the deepening of underwater shooting, the results obtained will be the darker the quality of the underwater image. Underwater imagery is widely used as an object in various activities such as underwater habitat mapping, underwater environment monitoring, underwater object search. Uneven lighting and colors that tend to be bluish and runny. One of the factors that influence the recognition results in pattern recognition is the quality of the image that is inputted. The image acquired from the source does not always have good quality. The process of improvement in digital images that experience interference in lighting and exposure to sunlight. The lighting repair process uses the retinex method and will compare the quality of the two methods later. Until later the results of both methods can be seen which is better. Retinex method can produce image improvement with high performance.Keywords: Digital Cintra, Underwater, Matlab Retinex Method

Sign in / Sign up

Export Citation Format

Share Document