Types of Defects - Food - Agriculture
Although this can sometimes signify that the produce is not ripe, it is also used as part of a grading process created for supermarkets. High-end supermarkets will favour perfect produce with target colour ranges / shades, and they are willing to pay more for it. So it’s in the farmers interest to ensure that it is graded accordingly and quickly.
Line scan colour cameras can help to automatically detect the produces’ colour and sort it at very high speeds – find out more about how they work.
Supermarkets set criteria for the size of the crops they require and farmers will want to ensure they have the biggest yield - reducing waste by only harvesting at the right times. Locating the position and size of the produce - while it’s in the field - can automate this process by generating data to help identify what should be picked (and when).
The location of each fruit or vegetable that meets the criteria can be passed onto an automated robotic picking system – reducing the need for manual harvesting. Any produce that isn’t picked is correlated with weather forecasts to help predict when the next optimum harvesting time would be.
All produce is naturally going to have some variation to its shape and the vision system will need to take this into account with its grading process.
The shape of produce is also used in the grading system used to find the “perfect” fruit & vegetables. So separating these from any odd-looking fruit and veg is also of interest to the farmers.
By using all image information from multiple image planes and sources (which can include 3D point cloud data from the LMI Gocator), it is able to identify the key features that help grade the produce.
Unfortunately defects are not always visible on the surface of the produce and techniques to be able to see beneath it need to be employed.
Delicate produce is susceptible to bruising during its journey from the plants to the supermarket shelves. Even with great care, there is still a chance that the operators and machinery involved during the harvesting and packaging processes can cause bruising – some of which may not be visible externally.
Mould can grow on living and decomposing produce and can spread pretty quickly. By the time its presence has been detected by the trained eye or colour vision systems, they have already formed large colonies. Early detection can help reduce contamination (and save costs) further down the line.
A lot of fruits are easy to identify as ripe simply by looking at their colour. Colour machine vision systems are perfect for this task. But how do you identify fruits that don’t change colour when they are ripe? Molecular level inspection is required to be able to detect these and the latest advances in machine vision technologies can help deliver this.
Pests & Disease
It is important to recognise early signs of pests and diseases in crops to be able to deal with the problem quickly. Generally, the diseases are identified and extracted manually, but a lot of the signs are not always visible on the surface of the crops and better inspection methods are required.
Recent advances in machine vision technology have introduced the ability to be able to see beyond the surface of fruit and vegetables to detect the “hidden” bruising, mould, disease, and pests that can not be detected manually or by colour machine vision systems. These systems can even detect how ripe a fruit is, which is especially useful for fruits that do not change colour when they are ripe (eg avocados).
Hyperspectral imaging enables molecular level inspection of organic products, bringing the advantages of spectroscopy into the machine vision arena.
Food and Beverage Solutions
Find out how machine vision can and has helped to improve automation and the efficiency of production processes in the food and beverage industries over the years.
Ask the expert!
Hi, I'm Stephan Nuijtemans, one of the sales specialists at STEMMER IMAGING.
Get in touch if you've got any questions on how to detect defects in food and we'll try our best to help you out.