首页> 外国专利> Computer implementation methods, computer program products, and computer processing systems for performing machine learning of image classification tasks, as well as advanced driver assistance systems for automobiles.

Computer implementation methods, computer program products, and computer processing systems for performing machine learning of image classification tasks, as well as advanced driver assistance systems for automobiles.

机译:用于执行图像分类任务的机器学习的计算机实现方法、计算机程序产品和计算机处理系统,以及用于汽车的高级驾驶员辅助系统。

摘要

PROBLEM TO BE SOLVED: To provide a data extension for an image classification task. A computer implementation method and system for performing machine learning of an image classification task are provided. This method involves selecting a first image and a second image from one or more training sets in one or more databases by a processor operably coupled to one or more databases. Including. This method uses a processor to average the intensities of each of a plurality of identical pixel pairs in the first image and the second image to form a mixed image with a second image. It further includes a step of superimposing images. This method further includes the step of training the machine learning process configured for the image classification task by using mixed images to extend the data used by the machine learning process of the image classification task by the processor. .. [Selection diagram] Fig. 3
机译:需要解决的问题:为图像分类任务提供数据扩展。本发明提供了一种用于执行图像分类任务的机器学习的计算机实现方法和系统。该方法涉及由可操作地耦合到一个或多个数据库的处理器从一个或多个数据库中的一个或多个训练集中选择第一图像和第二图像。包括…在内该方法使用处理器来平均第一图像和第二图像中的多个相同像素对中的每一个的强度,以形成与第二图像的混合图像。它还包括叠加图像的步骤。该方法还包括通过使用混合图像来训练为图像分类任务配置的机器学习过程的步骤,以扩展处理器用于图像分类任务的机器学习过程的数据。。[选择图]图3

著录项

  • 公开/公告号JP7034265B2

    专利类型

  • 公开/公告日2022-03-11

    原文格式PDF

  • 申请/专利权人 INTERNATIONAL BUSINESS MACHINES CORPORATION;

    申请/专利号JP20200514281

  • 发明设计人 井上 拓;

    申请日2018-09-20

  • 分类号G06T7;

  • 国家 JP

  • 入库时间 2022-08-24 23:58:53

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号