改進二維OTSU和自適應遺傳算法的紅外圖像分割
發(fā)布時間：2019-05-17 17:54
【摘要】：為提高電路板熱紅外圖像分割效果,解決電路板故障診斷的問題,提出了一種將自適應遺傳算法和改進二維OTSU算法相結合的新算法。自適應遺傳算法采取快速排序法,利用遺傳代數(shù)的變化改進交叉、變異概率;根據(jù)鄰域像素點與中心像素點之間的距離對二維OTSU鄰域均值算法的比例系數(shù)進行加權,判斷噪聲點并對熱紅外圖像進行降噪處理;引入類內方差法,對閾值進行優(yōu)化。利用了遺傳算法的并行性和強大的空間搜索能力,提高了二維OTSU的閾值查找速度,提高了熱紅外圖像的分割效率。實驗結果表明,該算法提高了熱紅外圖像的分割準確度,有一定的應用前景。
[Abstract]:In order to improve the thermal infrared image segmentation effect of circuit board and solve the problem of circuit board fault diagnosis, a new algorithm combining adaptive genetic algorithm with improved two-dimensional Otsu algorithm is proposed. The adaptive genetic algorithm adopts the fast sorting method to improve the cross and mutation probability by using the change of genetic algebra. According to the distance between neighborhood pixel and central pixel, the proportional coefficient of two-dimensional OTSU neighborhood mean algorithm is weighted, the noise point is judged and the thermal infrared image is de-noised, and the intra-class variance method is introduced to optimize the threshold. By using the parallelism of genetic algorithm and strong spatial search ability, the threshold search speed of two-dimensional Otsu is improved, and the segmentation efficiency of thermal infrared image is improved. The experimental results show that the algorithm improves the segmentation accuracy of thermal infrared image and has a certain application prospect.
【作者單位】： 中國民航大學;
【基金】：國家自然科學基金(U1333111,61405246) 中國民航大學中央高校項目(3122016D018)
【分類號】：TN21;TP391.41
本文編號：2479296
[Abstract]:In order to improve the thermal infrared image segmentation effect of circuit board and solve the problem of circuit board fault diagnosis, a new algorithm combining adaptive genetic algorithm with improved two-dimensional Otsu algorithm is proposed. The adaptive genetic algorithm adopts the fast sorting method to improve the cross and mutation probability by using the change of genetic algebra. According to the distance between neighborhood pixel and central pixel, the proportional coefficient of two-dimensional OTSU neighborhood mean algorithm is weighted, the noise point is judged and the thermal infrared image is de-noised, and the intra-class variance method is introduced to optimize the threshold. By using the parallelism of genetic algorithm and strong spatial search ability, the threshold search speed of two-dimensional Otsu is improved, and the segmentation efficiency of thermal infrared image is improved. The experimental results show that the algorithm improves the segmentation accuracy of thermal infrared image and has a certain application prospect.
【作者單位】： 中國民航大學;
【基金】：國家自然科學基金(U1333111,61405246) 中國民航大學中央高校項目(3122016D018)
【分類號】：TN21;TP391.41
【相似文獻】
相關期刊論文 前8條
1 官伯林;賈建援;朱應敏;;基于自適應遺傳算法的三軸光電跟蹤策略[J];儀器儀表學報;2012年08期
2 蘇琳琳;張曉林;;利用自適應遺傳算法的芯片功能驗證自動測試[J];應用科學學報;2011年06期
3 金力;劉橋;;基于自適應遺傳算法的運放的電路級綜合[J];西華大學學報(自然科學版);2006年02期
4 蘭海;汪宇涵;張利軍;;基于改進自適應遺傳算法的船舶電力系統(tǒng)濾波裝置優(yōu)化配置[J];船電技術;2009年06期
5 許川佩;陳征南;任智新;胡聰;;基于云自適應遺傳算法的NoC路徑分配研究[J];計算機測量與控制;2012年09期
6 許川佩;陳征南;任智新;;基于云自適應遺傳算法的NoC映射研究[J];計算機工程與應用;2012年36期
7 趙曙光,劉貴喜,楊萬海;利用自適應遺傳算法實現(xiàn)模擬電路自動設計[J];西安電子科技大學學報;2003年03期
8 ;[J];;年期
相關碩士學位論文 前3條
1 柯家偉;面向訂單快速交付的生產過程管控技術研究與系統(tǒng)實現(xiàn)[D];北京理工大學;2016年
2 陳殿夏;橋式網絡系統(tǒng)可靠性分析和優(yōu)化[D];沈陽工業(yè)大學;2005年
3 金力;基于改進自適應遺傳算法的CMOS運放的電路級綜合方法的研究[D];貴州大學;2006年
，本文編號：2479296
本文鏈接：http://sikaile.net/kejilunwen/dianzigongchenglunwen/2479296.html
