پديد آورندگان :
ﺟﻼﻟﯿﺎن ﺷﻬﺮي، ﻣﺤﻤﺪ رﺿﺎ داﻧﺸﮕﺎه ﻓﺮدوﺳﯽ ﻣﺸﻬﺪ - داﻧﺸﮑﺪه ﻣﻬﻨﺪﺳﯽ - ﮔﺮوه ﻣﻬﻨﺪﺳﯽ ﺑﺮق، ﻣﺸﻬﺪ، اﯾﺮان , ﻫﺎدي زاده، داﻧﺸﮕﺎه ﺻﻨﻌﺘﯽ ﻗﻮﭼﺎن - داﻧﺸﮑﺪه ﻣﻬﻨﺪﺳﯽ - ﮔﺮوه ﻣﻬﻨﺪﺳﯽ ﺑﺮق، ﻗﻮﭼﺎن، اﯾﺮان , ﺧﺎدﻣﯽ، ﻣﺮﺗﻀﯽ داﻧﺸﮕﺎه ﻓﺮدوﺳﯽ ﻣﺸﻬﺪ - داﻧﺸﮑﺪه ﻣﻬﻨﺪﺳﯽ - ﮔﺮوه ﻣﻬﻨﺪﺳﯽ ﺑﺮق، ﻣﺸﻬﺪ، اﯾﺮان , اﺑﺮاﻫﯿﻤﯽﻣﻘﺪم، ﻋﺒﺎس داﻧﺸﮕﺎه ﻓﺮدوﺳﯽ ﻣﺸﻬﺪ - داﻧﺸﮑﺪه ﻣﻬﻨﺪﺳﯽ - ﮔﺮوه ﻣﻬﻨﺪﺳﯽ ﺑﺮق، ﻣﺸﻬﺪ، اﯾﺮان
كليدواژه :
اﺳﺘﺨﺮاج وﯾﮋﮔﯽ , اﻟﮕﻮي دودوﯾﯽ ﻣﺤﻠﯽ , ﻃﺒﻘﻪﺑﻨﺪي ﺑﺎﻓﺖ , ﻧﻮﻓﻪ ﺳﻔﯿﺪ ﮔﻮﺳﯽ
چكيده فارسي :
ﻧﺨﺴﺘﯿﻦ ﮔﺎم در ﻃﺒﻘﻪﺑﻨﺪي ﺗﺼﺎوﯾﺮ ﺑﺎﻓﺘﯽ، ﺗﻮﺻﯿﻒ ﺑﺎﻓﺖ ﺑﺎ اﺳﺘﻔﺎده از اﺳﺘﺨﺮاج وﯾﮋﮔ ﯽﻫﺎي ﺗﺼﻮﯾﺮي ﻣﺨﺘﻠﻒ از آن اﺳﺖ. ﺗﺎﮐﻨﻮن روشﻫﺎي ﻣﺘﻌﺪدي ﺑﺮاي اﯾﻦ ﻣﻮﺿﻮع ﺗﻮﺳﻌﻪ ﯾﺎﻓﺘﻪاﻧﺪ ﮐﻪ از ﺟﻤﻠﻪ ﻣﺸﻬﻮرﺗﺮﯾﻦ آنﻫﺎ ﻣﯽﺗﻮان ﺑﻪ روش اﻟﮕﻮي دودوﯾﯽ ﻣﺤﻠﯽ اﺷﺎره ﮐﺮد. ﺑﻪﻣﻨﻈﻮر اﺳﺘﺨﺮاج اﻃﻼﻋﺎت ﺑﺎﻓﺘﯽ در ﻣﻘﯿﺎسﻫﺎي ﻣﺨﺘﻠﻒ، روش اﻟﮕﻮ ي ﺑﺎﯾﻨﺮي ﻣﺤﻠﯽ را ﻣﯽﺗﻮان در ﯾ ﮏ ﭼﻬﺎرﭼﻮب ﭼﻨﺪﻣﻘﯿﺎﺳﻪ ﭘﯿﺎدهﺳﺎزي ﮐﺮد. در اﯾﻦ ﺣﺎﻟﺖ، ﺑﺮدارﻫﺎي وﯾﮋﮔ ﯽ ﺑﻪدﺳﺖآﻣﺪه در ﺳﻄﻮح ﻣﻘﯿﺎس ﻣﺨﺘﻠﻒ ﺑﻪ ﯾﮑﺪﯾﮕﺮ ﭘﯿﻮﺳﺖ ﻣﯽﺷﻮﻧﺪ ﺗﺎ ﯾﮏ ﺑﺮدار وﯾﮋﮔ ﯽ ﺑﺮآﯾﻨﺪ ﺑﺎ ﻃﻮل ﺑﯿﺸﺘﺮ را ﺗﻮﻟﯿﺪ ﮐﻨﺪ؛ اﻣﺎ ﭼﻨﯿﻦ روﺷﯽ دو ﻋﯿﺐ ﻣﻬﻢ دارد؛ ﻧﺨﺴﺖاﯾﻦﮐﻪ، روش اﻟﮕﻮي دودوﯾﯽ ﻣﺤﻠ ﯽ ﺑﻪﺷﺪت ﻧﺴﺒﺖ ﺑﻪ ﻧﻮﻓﻪ ﺣﺴﺎس و ﺑﺎ اﻓﺰودن ﻧﻮﻓﻪ ﺑﻪ ﺗﺼﻮﯾﺮ ﺑﺎﻓﺘﯽ، ﺑﺮدارﻫﺎي وﯾﮋﮔ ﯽ ﺑﻪدﺳﺖآﻣﺪه ﻣﻤﮑﻦ اﺳﺖ ﺑﻪﺷﺪت ﺗﻐﯿﯿﺮ ﮐﻨﻨﺪ. دوماﯾﻦﮐﻪ، ﺑﺎ اﻓﺰاﯾﺶ ﺗﻌﺪاد ﻣﻘﯿﺎسﻫﺎ ، ﻃﻮل ﺑﺮدار وﯾﮋﮔ ﯽ ﺑﻪدﺳﺖآﻣﺪه ﻧﯿﺰ اﻓﺰاﯾﺶ ﻣﯽ ﯾﺎﺑﺪ ﮐﻪ اﯾﻦ اﻣﺮ ﺿﻤﻦ ﮐﺎﻫﺶ ﺳﺮﻋﺖ ﻓﺮآ ﯾﻨﺪ ﻃﺒﻘﻪﺑﻨﺪي ﺑﺎﻓﺖ، ﻣﻤﮑﻦ اﺳﺖ دﻗﺖ ﻃﺒﻘﻪﺑﻨﺪي را ﻧﯿﺰ ﮐﺎﻫﺶ دﻫﺪ. ﺑﺮاي رﻓﻊ و ﯾﺎ ﮐﺎﻫﺶ اﯾﻦ دو ﻋﯿﺐ ، در اﯾ ﻦ ﻣﻘﺎﻟﻪ، روﺷﯽ ﻣﺒﺘﻨﯽ ﺑﺮ اﻟﮕﻮي دودوﯾﯽ ﻣﺤﻠﯽ ﭼﻨﺪﻣﻘﯿﺎﺳﻪ ﭘﯿﺸﻨﻬﺎد ﻣﯽﺷﻮد ﮐﻪ از ﻣﻘﺎوﻣﺖ ﺑﻬﺘﺮ ي در ﻣﻘﺎﺑﻞ ﻧﻮﻓﻪ ﺳﻔﯿﺪ ﮔﻮﺳﯽ ﺑﺮﺧﻮردار و در ﻋﯿﻦ ﺣﺎل، ﻃﻮل ﺑﺮدار وﯾﮋﮔ ﯽ ﺗﻮﻟﯿﺪي ﺑﻪوﺳﯿﻠﮥ آن ﺑﻪﻃﻮردﻗﯿﻖ ﺑﺮاﺑﺮ ﺑﺎ ﻃﻮل ﺑﺮدار وﯾﮋﮔ ﯽ ﺗﻮﻟﯿﺪي ﺑﻪوﺳﯿﻠﮥ روش اﺻﻠﯽ اﻟﮕﻮي دودوﯾﯽ ﻣﺤﻠﯽ در ﺣﺎﻟﺖ ﺗﮏ ﻣﻘﯿﺎﺳﻪ اﺳﺖ. آزﻣﺎﯾﺶ ﻫﺎ ﺑﺮ روي ﭼﻬﺎر ﮔﺮوه از ﭘﺎﯾﮕﺎه داده Outexاﻧﺠﺎم ﺷﺪه ﮐﻪ آزﻣﺎﯾﺶﻫﺎي اﻧﺠﺎمﮔﺮﻓﺘﻪ ﻧﺸﺎندﻫﻨﺪه ﺑﺮﺗﺮي روش ﭘﯿﺸﻨﻬﺎدي ﻧﺴﺒﺖ ﺑﻪ روشﻫﺎي ﻣﻮﺟﻮد ﻣﺸﺎﺑﻪ اﺳﺖ.
چكيده لاتين :
In this paper we describe a novel noise-robust texture classification method using joint multiscale local binary pattern. The first step in texture classification is to describe the texture by extracting different features. So far, several methods have been developed for this topic, one of the most popular ones is Local Binary Pattern (LBP) method and its variants such as Completed Local Binary Pattern, Extended Local Binary Pattern, Local Temporary Pattern, Local Contrast Pattern, etc. In order to extract the features of a texture in different scales, the LBP method can be implemented in a multi-scale framework. For this purpose, the extracted feature vectors at different scales are usually concatenated together to produce the final feature vector with a longer length. But such a scheme has two main shortcomings. First, the LBP method is very sensitive to noise, hence by adding noise to a texture image, its feature vectors may change significantly. Second, by increasing the number of the scales, the length of the final feature vector is increased accordingly. This action increases the classification process time, and it may reduce the classification accuracy. To mitigate these shortcomings, this paper presents a method based on multiscale LBP, which has a better resistance against white Gaussian noise, while the length of its final feature vector is equal to the length of the final feature vector produced by the original LBP method. To implement the proposed method, we used 17 circular binary masks that contain 8 directed first-order masks, 8 directed second-order masks and 1 undirected mask. These masks have positive and negative weightes and each group of these masks have different radius which after convolution with input image extract features in different scales. Experiments were performed on four test groups of Outex database. Experimental results show that the proposed method is superior to the existing state-of-the-art methods. The complexity of proposed method is also analyzed. The results show that in this method, despite obtaining excellent classification accuracy, the complexity of the method has not changed much and even its complexity is less than some of the existing state-of-the-art methods.
