Fujitsu Laboratories Ltd. has announced that it is developing software that automatically generates image recognition programs that accurately detect the positions of components as they are captured by cameras in factory automation processes.
As it currently stands, automatically generated image processing programs have not been able to detect component positions, and as a result, individual image recognition programs must be developed. Fujitsu Laboratories was able to develop a technique for automatically generating image processing programs that detect positions by controlling the order in which the various functions that make up a program are combined, and using algorithms based on the similarity of shapes.
Automatically-generated image recognition programs use genetic algorithms. With these algorithms, two programs—which will become the "parents"—are randomly selected from multiple image recognition programs, known as parent generation. These parents are merged together to create a number of child image recognition programs, and each of these programs then undergoes an evaluation that uses training images and a set of target pictures, which show the features that are intended to be extracted from the pictures (the correct answer set).
Training pictures are fed into the child programs, and the output pictures are scored based on how well they match to the target images. The child programs are then culled based on their scores, with the highest-scoring programs becoming the parents to the next generation. Those that pass become the next parent generation, and the process is repeated until a child with a passing score appears. This method, however, can take a lot of time and cannot be used to automatically generate programs that will accurately detect the position of a component within an image.
Page 1 | Page 2
