Method for configuring an image analysis device, image analysis method, and image analysis device

Abstract

The aim of the invention is to configure an image analysis device (BA). This is achieved in that a plurality of training images (TPIC) assigned to an object type (OT) and an object sub-type (OST) are fed into a first neural network module (CNN) in order to detect image features. Furthermore, training output data sets (FEA) of the first neural network module (CNN) are fed into a second neural network module (MLP) in order to detect object types using image features. According to the invention, the first and second neural network module (CNN, MLP) are trained together such that training output data sets (OOT) of the second neural network module (MLP) at least approximately reproduce the object types (OT) assigned to the training images (TPIC). Furthermore, for each object type (OT1, OT2): - training images (TPIC) assigned to the object type (OT1, OT2) are fed into the trained first neural network module (CNN), - the first neural network module training output data set (FEA1, FEA2) generated for the respective training image (TPIC) is assigned to the object sub-type (OST) of the respective training image (TPIC), and - by means of the aforementioned sub-type assignments, a sub-type detection module (BMLP1. BMLP2) is configured to detect object sub-types (OST) using image features for the image analysis device (BA).

Description

Background technique [0001] For the automatic evaluation of image records, for example in medical diagnosis, in the monitoring of engineering or non-engineering systems and / or in the field of visual sensor systems for autonomous systems, machine learning methods are increasingly being used. With the help of this approach, learning-based image evaluation devices can be trained to automatically recognize objects displayed on images or associate them with object types. [0002] Thus, for example, an image evaluation device of a medical diagnostic device can be trained to specifically recognize a cell or tissue type on a microscopic image, or to associate with a cell type or a tissue type, respectively. In particular, the image evaluation device can be trained to associate image records of biological cells, such as blood cells, respectively, with a cell type as object type and with a developmental stage of this cell type as object subtype. [0003] A large number of preset images...

AI Technical Summary

Problems solved by technology

However, with the above approach, especially in the presence of rare object types, the required training overhead increases significantly

Images