基于大津和遺傳算法的紅外圖像分割-------外文翻譯(英文含翻譯).rar
基于大津和遺傳算法的紅外圖像分割-------外文翻譯(英文含翻譯),摘要:紅外圖像是由目標和環(huán)境的熱紅外射線形成的。隨著目標輻射出更多的熱量,我們能夠于端梢直方圖看到它們有著較大的灰度。根據(jù)紅外圖像的這個特性,這篇文章中提出了一種新的基于大津法和遺傳算法的紅外圖像分割方法。首先,將紅外圖像轉(zhuǎn)化為灰度圖像;其次,檢測圖像邊緣,并且通過大津和遺傳算法(otsu-ga)得出圖像分割的最佳閾值...
該文檔為壓縮文件,包含的文件列表如下:
內(nèi)容介紹
原文檔由會員 wanli1988go 發(fā)布
摘要:紅外圖像是由目標和環(huán)境的熱紅外射線形成的。隨著目標輻射出更多的熱量,我們能夠于端梢直方圖看到它們有著較大的灰度。根據(jù)紅外圖像的這個特性,這篇文章中提出了一種新的基于大津法和遺傳算法的紅外圖像分割方法。首先,將紅外圖像轉(zhuǎn)化為灰度圖像;其次,檢測圖像邊緣,并且通過大津和遺傳算法(Otsu-GA)得出圖像分割的最佳閾值;最后,根據(jù)閾值分割圖像,對比結(jié)果的雙峰和迭代。實驗結(jié)果顯示,該方法有效地提高了圖像分割的質(zhì)量。
關(guān)鍵詞:紅外圖像分割;閾值;大津算法;遺傳算法
Ⅰ 引言
紅外圖像分割在許多應(yīng)用中都很有用。從結(jié)果上看,可以在大場景中分別出興趣和目標區(qū)域,這對隨后的圖像分析或注釋十分有用。傳統(tǒng)的閾值分割方法有基于像素閾值,基于局部圖像閾值,基于過渡區(qū)域。然而,由于這個問題困難的本質(zhì),幾乎沒有自動的算法能夠在大量各種各樣數(shù)據(jù)中都能工作得很好[1-5]。
隨著目標輻射出更多的熱量,我們能夠于端梢直方圖看到它們有著較大的灰度,因為紅外圖像是由目標和環(huán)境的熱紅外射線形成的。根據(jù)紅外圖像的這個特性,這篇文章中提出了一種新的基于大津和遺傳算法的紅外圖像分割方法。
Abstract
¡ªThe infrared image is formed by the thermal infrared
rays of the targets and environment. As the targets radiate more
heat, we can see they have larger grey-scale, locating in the top-
end of histogram. According to this characteristic of infrared
images, a new method for infrared image segmentation based on
Otsu and genetic algorithm is proposed in this paper. First,
covert the infrared image to grayscale image; Second, detect the
image edge, and get the best threshold for image segmentation by
Otsu and GA (Otsu-GA); Finally, Segment the image by the
threshold, and contrast the results of two peaks and iteration.
Experimental results show that the method effectively improved
the quality of image segmentation.
Keywords-Infrared Image Segmentation; Threshold; Otsu;
Genetic Algorithm
關(guān)鍵詞:紅外圖像分割;閾值;大津算法;遺傳算法
Ⅰ 引言
紅外圖像分割在許多應(yīng)用中都很有用。從結(jié)果上看,可以在大場景中分別出興趣和目標區(qū)域,這對隨后的圖像分析或注釋十分有用。傳統(tǒng)的閾值分割方法有基于像素閾值,基于局部圖像閾值,基于過渡區(qū)域。然而,由于這個問題困難的本質(zhì),幾乎沒有自動的算法能夠在大量各種各樣數(shù)據(jù)中都能工作得很好[1-5]。
隨著目標輻射出更多的熱量,我們能夠于端梢直方圖看到它們有著較大的灰度,因為紅外圖像是由目標和環(huán)境的熱紅外射線形成的。根據(jù)紅外圖像的這個特性,這篇文章中提出了一種新的基于大津和遺傳算法的紅外圖像分割方法。
Abstract
¡ªThe infrared image is formed by the thermal infrared
rays of the targets and environment. As the targets radiate more
heat, we can see they have larger grey-scale, locating in the top-
end of histogram. According to this characteristic of infrared
images, a new method for infrared image segmentation based on
Otsu and genetic algorithm is proposed in this paper. First,
covert the infrared image to grayscale image; Second, detect the
image edge, and get the best threshold for image segmentation by
Otsu and GA (Otsu-GA); Finally, Segment the image by the
threshold, and contrast the results of two peaks and iteration.
Experimental results show that the method effectively improved
the quality of image segmentation.
Keywords-Infrared Image Segmentation; Threshold; Otsu;
Genetic Algorithm