Machine vision applications -  Successful examples from many industries

MACHINE VISION APPLICATIONS

Successful examples from many industries

Unusual imaging applications

February 2011

For years vision technology has been an established solution for optical control in the industrial arena. Is it possible to use this technology outside of the industrial field? The answer to this question is clearly "yes", as the following unusual examples show. These applications were realised by STEMMER IMAGING or partners over the years and indicate the possibilities that go beyond the standard applications.

Automated animal counting

Counting the population of animals is a necessary task in forestry to assure a well-balanced biota. As natural hunters such as bears or wolves became rare or extinct, men have to compensate by shooting the wild in order to control the population and to avoid excessive eating of the plants by deer or other animals. Precondition for this task is a prior registration of the wildlife stock, a task that is usually accomplished by the forest ranger that has to spend a long time on his raised hide.

How this time consuming task can be solved automatically with the aid of vision, has been shown by a clever forester in cooperation with STEMMER IMAGING using the software platform Common Vision Blox. He was able to develop a system that allowed him an assessment of the wildlife stock population despite the difficult environmental conditions such as changing lighting conditions, partially hidden objects and the different appearance in regard to size and colour. The imaging knowhow he gathered in this project was used years later in a similar system to count seagulls.

Even a count in water has already been successfully accomplished: An American fish farmer wanted to know how many young trouts are leaving the breeding basin at the end of the breeding season. This task seemed completely impossible in the beginning: to recognise and count moving, organic objects in water is not one of the standard functions for imaging systems. But even this task could be accomplished, based on components and knowhow from STEMMER IMAGING.

Animals that need to be counted can even be smaller: A well-know producer of insecticides used a vision system to detect plant louse nymphs and eggs in order to judge impact of his products.

"Head off!" using a vision system

A few years ago, a partner of STEMMER IMAGING in the Netherlands conducted a feasibility study to answer the question whether or not it is possible to process carrots automatically with the help of an imaging system and to determine the proper cutting edge for the greens. If too little is cut off, unwanted green stays, in cases where too much is cut off, this has a negative influence on the profitability of the process.

Even today harvesting and processing of vegetables is to a large extent done manually. Those tasks are not only monotonous and partly dangerous for the health; they also have a large impact on the price of the end product due to the high percentage of labour costs.

During the study washed carrots were put in arbitrary positions on a conveyor belt bypassing the imaging system. The image system had to recognise the rotation of the carrots, to trace the thickness along the direct axis, to detect the thickest position and to calculate the perfect cutting edge. The result of the study: Using appropriate hard- and software, it was possible to solve the task with the requested speed and precision.

This is how a burger should look!

The food industry is one of the areas of industry where the use of imaging systems has increased considerably during the last few years with a very wide range of applications. Some of the easier applications are the inspection of packages for food or drinks, but also the food stuff itself is more and more inspected and classified using vision.

There are applications where results gained from vision systems are used to calculate the weight and to add exact price markings. In some countries it is common that packages of meats and cold cuts have to contain exact portions of 100g. In order to get exactly 200 and not 212 grams of salami into a package, a customer of the UK subsidiary of STEMMER IMAGING uses image systems to recognise the percentage of fat and meat in the sausage slice. Considering the different specific weights, the results are used to control the thickness of the last slides that get into the package, in order to reach the exact target weight. In this application attention is also given to the fact that a slice with a low or high fat percentage is placed on top in the package in order to display an optically top quality product and to be able to adjust the price accordingly.

The visual appearance of a product plays also a major role for the big fast food chains. This leads to strange demands like some of them asking their suppliers to deliver their buns with an exactly defined browning degree and an approximately predefined amount and spread of the sesame seeds on top.

Imaging and vision for different application areas

As these examples show, imaging and vision is suitable for a variety of applications outside of the industrial field. Numerous additional exiting tasks have been solved successfully: Authenticity inspections of paintings, animation tasks in the entertainment industry, sports analysis, or intelligent traffic systems are only a few more examples for areas where the use of vision systems would not be expected. Certainly imaging cannot solve all tasks – but many more than commonly thought!

About STEMMER IMAGING

STEMMER IMAGING is Europe’s largest imaging technology provider and developer of the leading software platform Common Vision Blox. The company, which has its headquarters in Puchheim near Munich, provides its customers with all the components and services needed in order to develop reliable solutions for almost every industry sector. STEMMER IMAGING customers can benefit from a choice of machine vision products that is unique in Europe. These products from leading manufacturers are at the cutting edge of technology.