首页> 外国专利> IMAGE RECOGNITION BASED CELL RECOGNITION AND COUNTING METHOD AND APPARATUS, AND COMPUTER DEVICE

IMAGE RECOGNITION BASED CELL RECOGNITION AND COUNTING METHOD AND APPARATUS, AND COMPUTER DEVICE

机译:基于图像识别的小区识别和计数方法和装置,以及计算机设备

摘要

An image recognition based cell recognition and counting method and apparatus, and a computer device and a storage medium, relating to the technical field of artificial intelligence. The method comprises: acquiring a global digital cell image (202); selecting a regional image from the global digital cell image, wherein the regional image is a local image to be recognized and counted (204); carrying out nonlinear color transformation processing on the regional image to obtain a target image (206); and inputting the target image into a cell recognition and counting neural network which is trained on the basis of a historical global digital cell image containing unconventional cells, recognizing unconventional cells in the target image, and counting the recognized unconventional cells to obtain a cell recognition and counting result (208). The color of an image is made to be closer to a real image by means of nonlinear color transformation processing, and by using a trained cell recognition and counting network, unconventional cells in a target image can be quickly and accurately recognized and counted, thereby improving the efficiency of cell recognition and counting.
机译:基于图像识别的小区识别和计数方法和装置,以及与人工智能技术领域有关的计算机设备和存储介质。该方法包括:获取全局数字小区图像(202);从全局数字小区图像中选择区域图像,其中区域图像是要识别和计数的本地图像(204);在区域图像上进行非线性颜色变换处理以获得目标图像(206);并将目标图像输入到小区识别和计数神经网络,该神经网络基于包含非传统小区的历史全球数字小区图像训练,识别目标图像中的非传统小区,并计数识别的非传统细胞以获得细胞识别和计数结果(208)。通过非线性颜色变换处理使图像的颜色更靠近真实图像,并且通过使用培训的电池识别和数量网络,可以快速准确地识别和计算目标图像中的非传统小区,从而改善细胞识别与计数的效率。

著录项

  • 公开/公告号WO2021139258A1

    专利类型

  • 公开/公告日2021-07-15

    原文格式PDF

  • 申请/专利权人 PING AN TECHNOLOGY (SHENZHEN) CO. LTD.;

    申请/专利号WO2020CN118534

  • 发明设计人 GUO BINGXUE;LV CHUANFENG;

    申请日2020-09-28

  • 分类号G06T7;G06T7/11;G06T7/136;G06T5;

  • 国家 CN

  • 入库时间 2022-08-24 19:58:42

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