{ "source_file": "0930", "source_platform": "知乎专栏", "author": "stone", "title_listed": "图像语义分割准确率度量方法总结", "keywords": "语义分割,评价指标,mIoU", "domain": "图像处理", "content_type": "技术教程", "is_paid_flag": "否", "url": "https://zhuanlan.zhihu.com/p/38236530", "publish_date": "2018", "corpus_stage": "应用", "notes": "语义分割准确率度量方法总结", "id": "a11f96d519a0", "collected_at": "2026-10-02T18:43:39", "title": "图像语义分割准确率度量方法总结 - 知乎", "body_markdown": "[![](https://zhuanlan.zhihu.com/p/38236530)](javascript:void\\(0\\))\n![ZhiHu logo](../images/body-e1fa91dbdb48b8055184.png)\n[](https://www.zhihu.com)\n[关注](https://www.zhihu.com/signin?next=%2Ffollow)[推荐](https://www.zhihu.com/signin?next=%2F)[热榜](https://www.zhihu.com/signin?next=%2Fhot)[专栏](https://www.zhihu.com/signin?next=%2Fcolumn-square)[圈子](https://www.zhihu.com/signin?next=%2Fring-feeds)[AI WorksBeta ](https://www.zhihu.com/project-square)[故事](https://www.zhihu.com/fiore/h5/vip-web)\n​\n[直答](https://zhida.zhihu.com/)\n切换模式\n登录/注册\n图像语义分割准确率度量方法总结\n首发于[图像语义分割](https://www.zhihu.com/column/c_197474183)\n切换模式\n# 图像语义分割准确率度量方法总结\n[![stone](../images/011.jpg)](https://www.zhihu.com/people/big-big-stone)\n[stone](https://www.zhihu.com/people/big-big-stone)\n收藏之前点个赞呗\n[收录于 · 图像语义分割](https://www.zhihu.com/column/c_197474183)\n60 人赞同了该文章\n衡量图像语义分割准确率主要有三种方法：\n 1. 像素准确率（[pixel](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=pixel&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJwaXhlbCIsInpoaWRhX3NvdXJjZSI6ImVudGl0eSIsImNvbnRlbnRfaWQiOjc0ODYwNjgsImNvbnRlbnRfdHlwZSI6IkFydGljbGUiLCJtYXRjaF9vcmRlciI6MSwiemRfdG9rZW4iOm51bGx9.k6y5OWk_HiL9s7IYU0Iz08OCVB9nzoClKpS05tNOsw8&zhida_source=entity) accuracy, PA）\n 2. [平均像素准确率](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=%E5%B9%B3%E5%9D%87%E5%83%8F%E7%B4%A0%E5%87%86%E7%A1%AE%E7%8E%87&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLlubPlnYflg4_ntKDlh4bnoa7njociLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjEsInpkX3Rva2VuIjpudWxsfQ.WLT_3PtBgLvyiLIaqaZ9RRb-GySMJGSmJJaHX1t70RI&zhida_source=entity)（mean pixel accuracy, MPA）\n 3. 平均IOU（[Mean Intersection over Union](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=Mean+Intersection+over+Union&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJNZWFuIEludGVyc2VjdGlvbiBvdmVyIFVuaW9uIiwiemhpZGFfc291cmNlIjoiZW50aXR5IiwiY29udGVudF9pZCI6NzQ4NjA2OCwiY29udGVudF90eXBlIjoiQXJ0aWNsZSIsIm1hdGNoX29yZGVyIjoxLCJ6ZF90b2tlbiI6bnVsbH0.1q3nbK2Zn3D78lIEvAupvducCgsMBsRCQCnaG6Op_cQ&zhida_source=entity), MIOU ）\n\n\n在介绍三种方法之前，需要先说明一些符号表示的意义。\nkk ：类别总数，如果包括背景的话就是 k+1k+1\npijp_{ij} ：真实像素类别为 ii 的像素被预测为类别 jj 的总数量，换句话说，就是对于类别为 ii 的像素来说，被错分成类别 jj 的数量有多少。\nPiiP_{ii} ：真实像素类别为 ii 的像素被预测为类别 ii 的总数量，换句话说，就是对于真实类别为 ii 的像素来说，分对的像素总数有多少。 \n## 1.PA\nPA的意义很简单，和我们常规的分类准确率计算没有区别，就是把分对的像素总量除以像素总数。\n![](../images/001.jpg)\n## 2.MPA\nMPA是对PA的改进，它是先对每个类计算PA，然后再对所有类的PA求平均。\n![](../images/002.jpg)\n## 3.MIoU\n在[语义分割](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=2&q=%E8%AF%AD%E4%B9%89%E5%88%86%E5%89%B2&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLor63kuYnliIblibIiLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjIsInpkX3Rva2VuIjpudWxsfQ.mawh8KyynT7P_hzHK8qPEr80xFUeW2STZ9dY7lazyCs&zhida_source=entity)中，MIoU才是标准的准确率度量方法。它是分别对每个类计算（真实标签和预测结果的交并比）IOU，然后再对所有类别的IOU求均值。\n![](../images/003.jpg)\n编辑于 2018-06-19 13:07\n[深度学习（Deep Learning）](https://www.zhihu.com/topic/19813032)\n[机器学习](https://www.zhihu.com/topic/19559450)\n[图像分割](https://www.zhihu.com/topic/20137632)\n​赞同 60​​14 条评论​144 ​4 \n​分享\n​申请转载​\n​\n![](../images/004.jpeg)\n未登录用户\n \n\n14 条评论\n默认\n最新\n[![米小羊](../images/009.jpg)](https://www.zhihu.com/people/44841e91860dee150bc1f62e2be607c2)\n[米小羊](https://www.zhihu.com/people/44841e91860dee150bc1f62e2be607c2)\n请问有没有公式出处的文献？\n2022-03-02\n​回复​喜欢\n[![王子yi](../images/006.jpg)](https://www.zhihu.com/people/1550b5777e80254f9566b6bc15996cac)\n[王子yi](https://www.zhihu.com/people/1550b5777e80254f9566b6bc15996cac)\n你好，请问可以单独对一张图片计算miou吗\n2021-06-07\n​回复​喜欢\n展开其他 1 条回复​\n[![青绵鱼](../images/007.jpg)](https://www.zhihu.com/people/d477f0b64aba1a5bd668060bc3284beb)\n[青绵鱼](https://www.zhihu.com/people/d477f0b64aba1a5bd668060bc3284beb)\n请问有相关的代码实现么？\n2019-01-03\n​回复​喜欢\n展开其他 2 条回复​\n[![lvg锋](../images/008.jpg)](https://www.zhihu.com/people/6a1ac37cc3f5aab59b41e5ec7cf4a230)\n[lvg锋](https://www.zhihu.com/people/6a1ac37cc3f5aab59b41e5ec7cf4a230)\n你好，我想知道有k类的话，计算是否算上背景，因为如果算上背景的话，数值会提高不少\n2018-09-21\n​回复​喜欢\n展开其他 2 条回复​\n[![知乎用户pdl6D9](../images/009.jpg)](https://www.zhihu.com/people/fb128b56fdfed86d1a2c8237571d133f)\n[知乎用户pdl6D9](https://www.zhihu.com/people/fb128b56fdfed86d1a2c8237571d133f)\n想问一下，是对每张图片做统计算出某个度量标准了做平均还是将所有测试集图片汇总到一起做统计？\n2018-09-11\n​回复​喜欢\n展开其他 3 条回复​\n[![小小将](../images/010.jpg)](https://www.zhihu.com/people/fb31fbdb268e466a0928518a201696a9)\n[小小将](https://www.zhihu.com/people/fb31fbdb268e466a0928518a201696a9)\n​![](../images/body-2dd3e5c360100c555650.png)\n还有Dice Score，医学上常用\n2018-07-07\n​回复​3\n关于作者\n[![stone](../images/011.jpg)](https://www.zhihu.com/people/big-big-stone)\n[stone](https://www.zhihu.com/people/big-big-stone)\n收藏之前点个赞呗\n[ 回答 **30** ](https://www.zhihu.com/people/big-big-stone/answers)[ 文章 **15** ](https://www.zhihu.com/people/big-big-stone/posts)[ 关注者 **10,013** ](https://www.zhihu.com/people/big-big-stone/followers)\n​关注他​发私信\n大家都在搜\n换一换\n[江歌妈妈发长文尘埃终将落定704 万](https://www.zhihu.com/search?q=%E6%B1%9F%E6%AD%8C%E5%A6%88%E5%A6%88%E5%8F%91%E9%95%BF%E6%96%87%E5%B0%98%E5%9F%83%E7%BB%88%E5%B0%86%E8%90%BD%E5%AE%9A&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[中国男足 0-5 巴勒斯坦562 万](https://www.zhihu.com/search?q=%E4%B8%AD%E5%9B%BD%E7%94%B7%E8%B6%B3+0-5+%E5%B7%B4%E5%8B%92%E6%96%AF%E5%9D%A6&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[原央视主持人阿丘回应被通报419 万](https://www.zhihu.com/search?q=%E5%8E%9F%E5%A4%AE%E8%A7%86%E4%B8%BB%E6%8C%81%E4%BA%BA%E9%98%BF%E4%B8%98%E5%9B%9E%E5%BA%94%E8%A2%AB%E9%80%9A%E6%8A%A5&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[华为与赛力斯达成新五年合作409 万](https://www.zhihu.com/search?q=%E5%8D%8E%E4%B8%BA%E4%B8%8E%E8%B5%9B%E5%8A%9B%E6%96%AF%E8%BE%BE%E6%88%90%E6%96%B0%E4%BA%94%E5%B9%B4%E5%90%88%E4%BD%9C&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[迪拜航空确认航班发生事故398 万](https://www.zhihu.com/search?q=%E8%BF%AA%E6%8B%9C%E8%88%AA%E7%A9%BA%E7%A1%AE%E8%AE%A4%E8%88%AA%E7%8F%AD%E5%8F%91%E7%94%9F%E4%BA%8B%E6%95%85&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[东航回应网传空姐跪地道歉326 万](https://www.zhihu.com/search?q=%E4%B8%9C%E8%88%AA%E5%9B%9E%E5%BA%94%E7%BD%91%E4%BC%A0%E7%A9%BA%E5%A7%90%E8%B7%AA%E5%9C%B0%E9%81%93%E6%AD%89&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[王楚钦：不知道为什么就是感觉累320 万](https://www.zhihu.com/search?q=%E7%8E%8B%E6%A5%9A%E9%92%A6%EF%BC%9A%E4%B8%8D%E7%9F%A5%E9%81%93%E4%B8%BA%E4%BB%80%E4%B9%88%E5%B0%B1%E6%98%AF%E6%84%9F%E8%A7%89%E7%B4%AF&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[江苏高考作文《衬衫的价格为 9 磅 15 便士》爆火268 万](https://www.zhihu.com/search?q=%E6%B1%9F%E8%8B%8F%E9%AB%98%E8%80%83%E4%BD%9C%E6%96%87%E3%80%8A%E8%A1%AC%E8%A1%AB%E7%9A%84%E4%BB%B7%E6%A0%BC%E4%B8%BA+9+%E7%A3%85+15+%E4%BE%BF%E5%A3%AB%E3%80%8B%E7%88%86%E7%81%AB&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[女装网店开始用防拆带了265 万](https://www.zhihu.com/search?q=%E5%A5%B3%E8%A3%85%E7%BD%91%E5%BA%97%E5%BC%80%E5%A7%8B%E7%94%A8%E9%98%B2%E6%8B%86%E5%B8%A6%E4%BA%86&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[孕妇骑车别车被司机踹翻257 万](https://www.zhihu.com/search?q=%E5%AD%95%E5%A6%87%E9%AA%91%E8%BD%A6%E5%88%AB%E8%BD%A6%E8%A2%AB%E5%8F%B8%E6%9C%BA%E8%B8%B9%E7%BF%BB&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n### 推荐阅读\n[ ![mIoU及pixel accuracy ｜图像语义分割/语义图像生成中的评估指标 ｜ 公式解析附代码](../images/013.jpg)mIoU及pixel accuracy ｜图像语义分割/语义图像生成中的评估指标 ｜ 公式解析附代码 卡卡猡特 ](https://zhuanlan.zhihu.com/p/518797698)[ ![2019语义分割技术分享](../images/015.jpg)2019语义分割技术分享 小宇吱 ](https://zhuanlan.zhihu.com/p/74971350)[ ![语义分割概念及应用介绍](../images/017.jpg)语义分割概念及应用介绍 阿里云云栖号 ](https://zhuanlan.zhihu.com/p/46200875)[ ![语义分割中的深度学习方法全解：从FCN、SegNet到各代DeepLab](../images/019.jpg)语义分割中的深度学习方法全解：从FCN、SegNet到各代DeepLab 量子学园发表于量子位 ](https://zhuanlan.zhihu.com/p/27794982) \n\n_想来知乎工作？请发送邮件到 jobs@zhihu.com_\n打开知乎App\n在「我的页」右上角打开扫一扫\n![](../images/body-0f9788d006ab0f4da762.png)\n其他扫码方式：微信\n下载知乎App\n无障碍模式\n验证码登录\n密码登录\n[开通机构号](https://zhuanlan.zhihu.com/org/signup)\n中国 +86\n获取短信验证码\n忘记密码\n登录/注册\n其他方式登录\n未注册手机验证后自动登录，注册即代表同意[《知乎协议》](https://www.zhihu.com/term/zhihu-terms)[《隐私保护指引》](https://www.zhihu.com/term/privacy)\n扫码下载知乎 App\n关闭二维码\n", "metadata": { "title": "图像语义分割准确率度量方法总结 - 知乎", "description": "衡量图像语义分割准确率主要有三种方法： 像素准确率（pixel accuracy, PA）平均像素准确率（mean pixel accuracy, MPA）平均IOU（Mean Intersection over Union, MIOU ）在介绍三种方法之前，需要先说明一些符号…", "keywords": "深度学习（Deep Learning）,机器学习,图像分割", "author": null, "og:title": "图像语义分割准确率度量方法总结", "og:url": "https://zhuanlan.zhihu.com/p/38236530", "og:description": "衡量图像语义分割准确率主要有三种方法： 像素准确率（pixel accuracy, PA）平均像素准确率（mean pixel accuracy, MPA）平均IOU（Mean Intersection over Union, MIOU ）在介绍三种方法之前，需要先说明一些符号…", "og:type": "article", "og:site_name": "知乎专栏" }, "paywall": { "listed_paid": false, "detected": false, "marker": "", "public_only": false, "policy": "付费/会员内容只保留页面公开可见部分，不绕过付费墙" }, "crawl": { "success": true, "status_code": 200, "error": "" }, "code_blocks": [], "images": [ { "status": "http_403", "url": "https://zhuanlan.zhihu.com/p/38236530", "src": "https://zhuanlan.zhihu.com/p/38236530", "alt": "", "downloaded": false, "discovery": "saved_markdown", "retried_at": "2026-10-20T17:45:47Z" } ], "audio": [], "videos": [], "links": [ { "type": "internal", "href": "https://www.zhihu.com" }, { "type": "internal", "href": "https://www.zhihu.com/signin?next=%2Ffollow" }, { "type": "internal", "href": "https://www.zhihu.com/signin?next=%2F" }, { "type": "internal", "href": "https://www.zhihu.com/signin?next=%2Fhot" }, { "type": "internal", "href": "https://www.zhihu.com/signin?next=%2Fcolumn-square" }, { "type": "internal", "href": "https://www.zhihu.com/signin?next=%2Fring-feeds" }, { "type": "internal", "href": "https://www.zhihu.com/project-square" }, { "type": "internal", "href": "https://www.zhihu.com/fiore/h5/vip-web" }, { "type": "internal", "href": "https://zhida.zhihu.com/" }, { "type": "internal", "href": "https://www.zhihu.com/column/c_197474183" }, { "type": "internal", "href": "https://www.zhihu.com/people/big-big-stone" }, { "type": "internal", "href": "https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=pixel&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJwaXhlbCIsInpoaWRhX3NvdXJjZSI6ImVudGl0eSIsImNvbnRlbnRfaWQiOjc0ODYwNjgsImNvbnRlbnRfdHlwZSI6IkFydGljbGUiLCJtYXRjaF9vcmRlciI6MSwiemRfdG9rZW4iOm51bGx9.k6y5OWk_HiL9s7IYU0Iz08OCVB9nzoClKpS05tNOsw8&zhida_source=entity" }, { "type": "internal", "href": "https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=%E5%B9%B3%E5%9D%87%E5%83%8F%E7%B4%A0%E5%87%86%E7%A1%AE%E7%8E%87&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLlubPlnYflg4_ntKDlh4bnoa7njociLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjEsInpkX3Rva2VuIjpudWxsfQ.WLT_3PtBgLvyiLIaqaZ9RRb-GySMJGSmJJaHX1t70RI&zhida_source=entity" }, { "type": "internal", "href": "https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=Mean+Intersection+over+Union&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJNZWFuIEludGVyc2VjdGlvbiBvdmVyIFVuaW9uIiwiemhpZGFfc291cmNlIjoiZW50aXR5IiwiY29udGVudF9pZCI6NzQ4NjA2OCwiY29udGVudF90eXBlIjoiQXJ0aWNsZSIsIm1hdGNoX29yZGVyIjoxLCJ6ZF90b2tlbiI6bnVsbH0.1q3nbK2Zn3D78lIEvAupvducCgsMBsRCQCnaG6Op_cQ&zhida_source=entity" }, { "type": "internal", "href": "https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=2&q=%E8%AF%AD%E4%B9%89%E5%88%86%E5%89%B2&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLor63kuYnliIblibIiLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjIsInpkX3Rva2VuIjpudWxsfQ.mawh8KyynT7P_hzHK8qPEr80xFUeW2STZ9dY7lazyCs&zhida_source=entity" }, { "type": "internal", "href": "https://www.zhihu.com/topic/19813032" }, { "type": "internal", "href": "https://www.zhihu.com/topic/19559450" }, { "type": "internal", "href": "https://www.zhihu.com/topic/20137632" }, { "type": "internal", "href": "https://www.zhihu.com/people/44841e91860dee150bc1f62e2be607c2" }, { "type": "internal", "href": "https://www.zhihu.com/people/1550b5777e80254f9566b6bc15996cac" }, { "type": "external", "href": "javascript:void(0)" } ], "repair_version": 2, "repaired_at": "2026-10-20T17:26:12Z", "source_url": "https://zhuanlan.zhihu.com/p/38236530", "delivery_status": "unverified", "discovery_status": "http_403", "missing_reason": "源页面未重新核验:http_403；已按保留正文补采图片", "repair_error": "", "actual_video_files": [], "expected_saved_body_images": 1, "image_discovery": "saved_markdown", "source_body_markdown": "[![](https://zhuanlan.zhihu.com/p/38236530)](javascript:void\\(0\\))\n![ZhiHu logo](https://static.zhihu.com/heifetz/assets/wechat-share-logo.39ea9ecd.png)\n[](https://www.zhihu.com)\n[关注](https://www.zhihu.com/signin?next=%2Ffollow)[推荐](https://www.zhihu.com/signin?next=%2F)[热榜](https://www.zhihu.com/signin?next=%2Fhot)[专栏](https://www.zhihu.com/signin?next=%2Fcolumn-square)[圈子](https://www.zhihu.com/signin?next=%2Fring-feeds)[AI WorksBeta ](https://www.zhihu.com/project-square)[故事](https://www.zhihu.com/fiore/h5/vip-web)\n​\n[直答](https://zhida.zhihu.com/)\n切换模式\n登录/注册\n图像语义分割准确率度量方法总结\n首发于[图像语义分割](https://www.zhihu.com/column/c_197474183)\n切换模式\n# 图像语义分割准确率度量方法总结\n[![stone](https://picx.zhimg.com/36f310418_l.jpg?source=32738c0c&needBackground=1)](https://www.zhihu.com/people/big-big-stone)\n[stone](https://www.zhihu.com/people/big-big-stone)\n收藏之前点个赞呗\n[收录于 · 图像语义分割](https://www.zhihu.com/column/c_197474183)\n60 人赞同了该文章\n衡量图像语义分割准确率主要有三种方法：\n 1. 像素准确率（[pixel](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=pixel&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJwaXhlbCIsInpoaWRhX3NvdXJjZSI6ImVudGl0eSIsImNvbnRlbnRfaWQiOjc0ODYwNjgsImNvbnRlbnRfdHlwZSI6IkFydGljbGUiLCJtYXRjaF9vcmRlciI6MSwiemRfdG9rZW4iOm51bGx9.k6y5OWk_HiL9s7IYU0Iz08OCVB9nzoClKpS05tNOsw8&zhida_source=entity) accuracy, PA）\n 2. [平均像素准确率](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=%E5%B9%B3%E5%9D%87%E5%83%8F%E7%B4%A0%E5%87%86%E7%A1%AE%E7%8E%87&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLlubPlnYflg4_ntKDlh4bnoa7njociLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjEsInpkX3Rva2VuIjpudWxsfQ.WLT_3PtBgLvyiLIaqaZ9RRb-GySMJGSmJJaHX1t70RI&zhida_source=entity)（mean pixel accuracy, MPA）\n 3. 平均IOU（[Mean Intersection over Union](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=1&q=Mean+Intersection+over+Union&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiJNZWFuIEludGVyc2VjdGlvbiBvdmVyIFVuaW9uIiwiemhpZGFfc291cmNlIjoiZW50aXR5IiwiY29udGVudF9pZCI6NzQ4NjA2OCwiY29udGVudF90eXBlIjoiQXJ0aWNsZSIsIm1hdGNoX29yZGVyIjoxLCJ6ZF90b2tlbiI6bnVsbH0.1q3nbK2Zn3D78lIEvAupvducCgsMBsRCQCnaG6Op_cQ&zhida_source=entity), MIOU ）\n\n\n在介绍三种方法之前，需要先说明一些符号表示的意义。\nkk ：类别总数，如果包括背景的话就是 k+1k+1\npijp_{ij} ：真实像素类别为 ii 的像素被预测为类别 jj 的总数量，换句话说，就是对于类别为 ii 的像素来说，被错分成类别 jj 的数量有多少。\nPiiP_{ii} ：真实像素类别为 ii 的像素被预测为类别 ii 的总数量，换句话说，就是对于真实类别为 ii 的像素来说，分对的像素总数有多少。 \n## 1.PA\nPA的意义很简单，和我们常规的分类准确率计算没有区别，就是把分对的像素总量除以像素总数。\n![](https://pic2.zhimg.com/v2-b3fdc8f9c3719751a1d90c8a5abf7d77_1440w.jpg)\n## 2.MPA\nMPA是对PA的改进，它是先对每个类计算PA，然后再对所有类的PA求平均。\n![](https://pica.zhimg.com/v2-c95ad8df0570cc033b8910a9d66d5e3e_1440w.jpg)\n## 3.MIoU\n在[语义分割](https://zhida.zhihu.com/search?content_id=7486068&content_type=Article&match_order=2&q=%E8%AF%AD%E4%B9%89%E5%88%86%E5%89%B2&zd_token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJ6aGlkYV9zZXJ2ZXIiLCJleHAiOjE3OTExMzk0MTksInEiOiLor63kuYnliIblibIiLCJ6aGlkYV9zb3VyY2UiOiJlbnRpdHkiLCJjb250ZW50X2lkIjo3NDg2MDY4LCJjb250ZW50X3R5cGUiOiJBcnRpY2xlIiwibWF0Y2hfb3JkZXIiOjIsInpkX3Rva2VuIjpudWxsfQ.mawh8KyynT7P_hzHK8qPEr80xFUeW2STZ9dY7lazyCs&zhida_source=entity)中，MIoU才是标准的准确率度量方法。它是分别对每个类计算（真实标签和预测结果的交并比）IOU，然后再对所有类别的IOU求均值。\n![](https://picx.zhimg.com/v2-fb4236ce3bb23dc6f35802878493243b_1440w.jpg)\n编辑于 2018-06-19 13:07\n[深度学习（Deep Learning）](https://www.zhihu.com/topic/19813032)\n[机器学习](https://www.zhihu.com/topic/19559450)\n[图像分割](https://www.zhihu.com/topic/20137632)\n​赞同 60​​14 条评论​144 ​4 \n​分享\n​申请转载​\n​\n![](https://pic1.zhimg.com/v2-d41c2ceaed8f51999522f903672a521f_l.jpeg)\n未登录用户\n \n\n14 条评论\n默认\n最新\n[![米小羊](https://picx.zhimg.com/v2-abed1a8c04700ba7d72b45195223e0ff_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/44841e91860dee150bc1f62e2be607c2)\n[米小羊](https://www.zhihu.com/people/44841e91860dee150bc1f62e2be607c2)\n请问有没有公式出处的文献？\n2022-03-02\n​回复​喜欢\n[![王子yi](https://pica.zhimg.com/v2-8945402d6a8c67b7664a180454321f5e_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/1550b5777e80254f9566b6bc15996cac)\n[王子yi](https://www.zhihu.com/people/1550b5777e80254f9566b6bc15996cac)\n你好，请问可以单独对一张图片计算miou吗\n2021-06-07\n​回复​喜欢\n展开其他 1 条回复​\n[![青绵鱼](https://picx.zhimg.com/ac2f529f9a73559c207821f634974272_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/d477f0b64aba1a5bd668060bc3284beb)\n[青绵鱼](https://www.zhihu.com/people/d477f0b64aba1a5bd668060bc3284beb)\n请问有相关的代码实现么？\n2019-01-03\n​回复​喜欢\n展开其他 2 条回复​\n[![lvg锋](https://pica.zhimg.com/v2-4c7cf3aa7c3efb0434a1c0c84543d194_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/6a1ac37cc3f5aab59b41e5ec7cf4a230)\n[lvg锋](https://www.zhihu.com/people/6a1ac37cc3f5aab59b41e5ec7cf4a230)\n你好，我想知道有k类的话，计算是否算上背景，因为如果算上背景的话，数值会提高不少\n2018-09-21\n​回复​喜欢\n展开其他 2 条回复​\n[![知乎用户pdl6D9](https://picx.zhimg.com/v2-abed1a8c04700ba7d72b45195223e0ff_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/fb128b56fdfed86d1a2c8237571d133f)\n[知乎用户pdl6D9](https://www.zhihu.com/people/fb128b56fdfed86d1a2c8237571d133f)\n想问一下，是对每张图片做统计算出某个度量标准了做平均还是将所有测试集图片汇总到一起做统计？\n2018-09-11\n​回复​喜欢\n展开其他 3 条回复​\n[![小小将](https://pic1.zhimg.com/v2-4c580ad38bc656abf65b1a7fb14d4573_l.jpg?source=06d4cd63)](https://www.zhihu.com/people/fb31fbdb268e466a0928518a201696a9)\n[小小将](https://www.zhihu.com/people/fb31fbdb268e466a0928518a201696a9)\n​![](https://picx.zhimg.com/v2-27bfcba90e66db79ce8768ab807e017e_l.png?source=32738c0c)\n还有Dice Score，医学上常用\n2018-07-07\n​回复​3\n关于作者\n[![stone](https://picx.zhimg.com/36f310418_l.jpg?source=32738c0c&needBackground=1)](https://www.zhihu.com/people/big-big-stone)\n[stone](https://www.zhihu.com/people/big-big-stone)\n收藏之前点个赞呗\n[ 回答 **30** ](https://www.zhihu.com/people/big-big-stone/answers)[ 文章 **15** ](https://www.zhihu.com/people/big-big-stone/posts)[ 关注者 **10,013** ](https://www.zhihu.com/people/big-big-stone/followers)\n​关注他​发私信\n大家都在搜\n换一换\n[江歌妈妈发长文尘埃终将落定704 万](https://www.zhihu.com/search?q=%E6%B1%9F%E6%AD%8C%E5%A6%88%E5%A6%88%E5%8F%91%E9%95%BF%E6%96%87%E5%B0%98%E5%9F%83%E7%BB%88%E5%B0%86%E8%90%BD%E5%AE%9A&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[中国男足 0-5 巴勒斯坦562 万](https://www.zhihu.com/search?q=%E4%B8%AD%E5%9B%BD%E7%94%B7%E8%B6%B3+0-5+%E5%B7%B4%E5%8B%92%E6%96%AF%E5%9D%A6&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[原央视主持人阿丘回应被通报419 万](https://www.zhihu.com/search?q=%E5%8E%9F%E5%A4%AE%E8%A7%86%E4%B8%BB%E6%8C%81%E4%BA%BA%E9%98%BF%E4%B8%98%E5%9B%9E%E5%BA%94%E8%A2%AB%E9%80%9A%E6%8A%A5&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[华为与赛力斯达成新五年合作409 万](https://www.zhihu.com/search?q=%E5%8D%8E%E4%B8%BA%E4%B8%8E%E8%B5%9B%E5%8A%9B%E6%96%AF%E8%BE%BE%E6%88%90%E6%96%B0%E4%BA%94%E5%B9%B4%E5%90%88%E4%BD%9C&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[迪拜航空确认航班发生事故398 万](https://www.zhihu.com/search?q=%E8%BF%AA%E6%8B%9C%E8%88%AA%E7%A9%BA%E7%A1%AE%E8%AE%A4%E8%88%AA%E7%8F%AD%E5%8F%91%E7%94%9F%E4%BA%8B%E6%95%85&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[东航回应网传空姐跪地道歉326 万](https://www.zhihu.com/search?q=%E4%B8%9C%E8%88%AA%E5%9B%9E%E5%BA%94%E7%BD%91%E4%BC%A0%E7%A9%BA%E5%A7%90%E8%B7%AA%E5%9C%B0%E9%81%93%E6%AD%89&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[王楚钦：不知道为什么就是感觉累320 万](https://www.zhihu.com/search?q=%E7%8E%8B%E6%A5%9A%E9%92%A6%EF%BC%9A%E4%B8%8D%E7%9F%A5%E9%81%93%E4%B8%BA%E4%BB%80%E4%B9%88%E5%B0%B1%E6%98%AF%E6%84%9F%E8%A7%89%E7%B4%AF&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)热\n[江苏高考作文《衬衫的价格为 9 磅 15 便士》爆火268 万](https://www.zhihu.com/search?q=%E6%B1%9F%E8%8B%8F%E9%AB%98%E8%80%83%E4%BD%9C%E6%96%87%E3%80%8A%E8%A1%AC%E8%A1%AB%E7%9A%84%E4%BB%B7%E6%A0%BC%E4%B8%BA+9+%E7%A3%85+15+%E4%BE%BF%E5%A3%AB%E3%80%8B%E7%88%86%E7%81%AB&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[女装网店开始用防拆带了265 万](https://www.zhihu.com/search?q=%E5%A5%B3%E8%A3%85%E7%BD%91%E5%BA%97%E5%BC%80%E5%A7%8B%E7%94%A8%E9%98%B2%E6%8B%86%E5%B8%A6%E4%BA%86&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n[孕妇骑车别车被司机踹翻257 万](https://www.zhihu.com/search?q=%E5%AD%95%E5%A6%87%E9%AA%91%E8%BD%A6%E5%88%AB%E8%BD%A6%E8%A2%AB%E5%8F%B8%E6%9C%BA%E8%B8%B9%E7%BF%BB&search_source=Trending&utm_content=search_hot&utm_medium=organic&utm_source=zhihu&type=content&request_click_id=13835657-9306-4244-b2d2-3f1a23781479)\n### 推荐阅读\n[ ![mIoU及pixel accuracy ｜图像语义分割/语义图像生成中的评估指标 ｜ 公式解析附代码](https://picx.zhimg.com/v2-49b7524944d0c6db80424fe460693876_250x0.jpg?source=172ae18b)mIoU及pixel accuracy ｜图像语义分割/语义图像生成中的评估指标 ｜ 公式解析附代码 卡卡猡特 ](https://zhuanlan.zhihu.com/p/518797698)[ ![2019语义分割技术分享](https://pica.zhimg.com/v2-28ed9a236a24b47d3e6312c4c793aa3c_250x0.jpg?source=172ae18b)2019语义分割技术分享 小宇吱 ](https://zhuanlan.zhihu.com/p/74971350)[ ![语义分割概念及应用介绍](https://picx.zhimg.com/v2-15d9e439be7dffea4c7d791b83912f96_250x0.jpg?source=172ae18b)语义分割概念及应用介绍 阿里云云栖号 ](https://zhuanlan.zhihu.com/p/46200875)[ ![语义分割中的深度学习方法全解：从FCN、SegNet到各代DeepLab](https://picx.zhimg.com/v2-aa833af944437d9ce2a4e853c93dda56_250x0.jpg?source=172ae18b)语义分割中的深度学习方法全解：从FCN、SegNet到各代DeepLab 量子学园发表于量子位 ](https://zhuanlan.zhihu.com/p/27794982) \n\n_想来知乎工作？请发送邮件到 jobs@zhihu.com_\n打开知乎App\n在「我的页」右上角打开扫一扫\n![](https://picx.zhimg.com/v2-9e41ea16bdfbe9cf4896617ecad5b4ca.png)\n其他扫码方式：微信\n下载知乎App\n无障碍模式\n验证码登录\n密码登录\n[开通机构号](https://zhuanlan.zhihu.com/org/signup)\n中国 +86\n获取短信验证码\n忘记密码\n登录/注册\n其他方式登录\n未注册手机验证后自动登录，注册即代表同意[《知乎协议》](https://www.zhihu.com/term/zhihu-terms)[《隐私保护指引》](https://www.zhihu.com/term/privacy)\n扫码下载知乎 App\n关闭二维码\n", "body_base_path": "attachments/a11f96d519a0/pages/", "body_markdown_path": "attachments/a11f96d519a0/pages/archived-body.md", "local_html": "attachments/a11f96d519a0/pages/archived-body.html", "body_source": "prior_capture" }