Image Processing

Automated extraction and analysis of vast amounts of actionable information

Image processing enables all kinds of useful information to be extracted from images or video for applications ranging from medical diagnosis, robotics and manufacturing quality control to surveillance and security systems.

The images don’t have to be traditional colour or black-and-white images – they can be gathered using techniques such as thermal imaging, laser scanning or 3D depth imaging.

The key benefit is that image processing automates the extraction of useful and actionable insights from imagery.

This could involve monitoring congestion as part of a smart traffic system to warn drivers which roads to avoid, for example, or analysing CCTV images as part of a security system and raising the alarm if a perimeter has been breached.

Real-world challenges

Removing major bottleneck for Above Surveying

Maximising the efficiency of solar farms is critical for operators, with drones often used to capture precise thermographic and video diagnostic data to warn of any defects.

When Above Surveying approached Plextek for help with automating its labour-intensive post-survey work, the challenge was to translate its data into a format that machine learning tools could process – removing solar glint along the way.

We recommended five areas of investigation along with example processing chains which could improve the client’s current processing workflow to increase overall efficiency and add extra capabilities.

Pairing ML with semantic segmentation

The challenge from the client was to take image retrieval to the next level – improving the performance of traditional visual scene mapping (VSM) algorithms when it comes to finding the best match to an image from a dataset of georeferenced images.

The Plextek team decided to augment the traditional algorithms with the latest machine learning techniques. Semantic segmentation was used to assign a label to each pixel in an image, based on which category it belonged to – tree, person, vehicle, etc. This allowed the image matching algorithms to focus on features that are likely to remain consistent over time (such as buildings) and to ignore features which are likely to change (e.g. people, vehicles).

As traditional VSM algorithms were still performing the image retrieval, desirable properties such as explainability and consistency were retained – but with the added benefit of the improved performance offered using machine learning.

Key skills

  • Sensor modalities

    Not all cameras produce standard colour imagery and instead thermal camera, depth cameras or even acoustic images may be more suited to the application.

  • Camera choice

    When selecting the optimum camera, it is necessary to consider various parameters, including the number of pixels, the field of view and the frame rate.

  • Algorithm development

    Neural networks and other machine learning algorithms can be used to achieve close to human-level performance in object recognition and other tasks.

  • Embedded processing

    To reduce the bandwidth required to transfer raw data from a camera, embedded processing is used to extract and transmit valuable information instead.

Video from cameras all around us is ubiquitous in the modern world but it is easy for humas to become inundated with data. Image processing is therefore necessary to automatically reduce this raw data deluge to a more manageable stream of valuable information.

Dr Damien Clarke, Lead Consultant
Dr Damien Clarke

Lead Consultant

What sets us apart when it comes to image processing?

Extracting information from a video stream is the key requirement when developing image-processing-based solutions.

Plextek collect test data to support the development and implementation of a suitable algorithmic solution. It is not just the image processing step itself that is important when producing an image processing solution for a client; we have expertise in the entire end-to-end development process, including:

  • Camera selection
  • Algorithm development
  • Hardware design
  • Embedded processing
  • Manufacturing

An additional area of expertise is the design of bespoke hardware systems for performing embedded processing, which includes ML using specialised hardware. This covers areas such as:

Contact Plextek

Contact Us

Got a question?

If you have got a question, or even just an idea, get in touch

Projects

View All
Machine Learning for Rapid Propagation Assessment
Machine Learning for Rapid Propagation Assessment

Developing a groundbreaking ML model for swift and efficient coverage prediction in complex urban environments, enabling rapid optimisation of transmitter locations on standard computing hardware.

Read More
Immersive Technology for Complex Systems Training
Immersive Technology for Complex Systems Training

using the latest immersive technologies to deliver realistic and engaging training solutions for complex systems, ensuring specialist procedures are effectively imparted and measured within any environment.

Read More
Drone Inspection of Solar Farms
Drone Inspection of Solar Farms

Enhancing solar farm efficiency through advanced drone inspection and automated analysis of diagnostic data.

Read More
Microwave Camera for Detecting Hidden Objects
Microwave Camera for Detecting Hidden Objects

Developing a groundbreaking microwave camera technology, enabling real-time detection of concealed items in a non-intrusive manner, improving airport security with minimal passenger disruption.

Read More