The drone industry has evolved remarkably in the past few years, with a vast number of companies emerging from this exciting revolution. To make things even more attractive, civil aviation authorities from several countries (including Australia) are opting for less restrictive rules when it comes down to flying sub-2Kg aircraft not above 400 feet.

We support this move as a necessary stage of the maturity process that the RPAS industry is undergoing. It will ultimately strengthen the drone services market, as only the most experienced, safety-oriented and reliable operators will make their way through to the top. However, we are very concerned about unprofessional services being delivered in the interim, blurring the benefits that the RPAS industry can offer.

We were recently approached by a client who had engaged with an operator with little experience with complex operations. He was really disappointed with the results obtained for a project with the following requirements:

• Survey 3.7Km of water canals, to map the embankments between -0.6m and 1.6m (MSL);
• Lay and measure ground control points in the area, to ensure high accuracy
• Data collection on low tides was a strong requirement, as it was necessary to maximise the exposure of the formed banks;
• Conduct safe and compliant flights over a densely populated neighbourhood;
• Generate a Digital Terrain Model (DTM) of the area, removing all built structures over water and connected to the banks, such as jetties and pontoons, and all stationary watercraft;
• Generate a high-definition orthomosaic of area.

Obviously, a challenging project. After all, there are 160 pontoons in that area. Yet, it is totally achievable if the right software, techniques and skills are employed.

We happened to have access to part of the deliverables previously sent to the client, and the results speak for themselves:

In a project requiring proper engineering rigor, delivering a DTM in which vegetation is still influencing surface modelling is definitely not a good sign.

Scout Aerial engaged to conduct all new flight operations and imagery processing steps. Not even the ground control point marks laid by the former operator could be reused, as they were too small for proper imagery georeferencing. The image below shows some GCP measurement locations (white crosses). You can probably see one straight away (laid by Scout Aerial surveyors). Now try to find the other ones laid by the former operator:

Interested to see the difference? See below some samples of our work:

Do you have a similar problem? Contact us now and have it solved the right way.

It is no news that remotely piloted aircraft are revolutionising the way companies are undertaking periodic inspections on their critical assets. Gone are the days when people had to risk their lives approaching high-voltage power lines, climbing 100-meters high wind turbines or hanging on a bridge over dangerous waters, trying to locate missing bolts, cracks, water leaks and rust, while trying to balance risk factors and minimising safety hazards.

Scout Aerial is part of this revolution, delivering extremely detailed asset inspections, combining the benefits of aerial images, videos and 3D modelling, thanks to our:

• State-of-the-art RPAS and specialised sensors (including LiDAR and thermal cameras), dual-screen visualisation and GNSS-based navigation;
• Highly trained and certified RPA pilots;
• Experienced 3D modelling and GIS specialists;
• Powerful graphic processing servers.

We had the opportunity to partner with one of the main water supply and sewerage service providers operating in Queensland, in a project to inspect the main assets in one of its facilities: a water tower and a reservoir. Although located in a relatively small area, the inspected site required careful planning, due to unique characteristics:

• The water tower contained several RF antennas and transmitters installed around its structure. They literally created an invisible shield around 20% of its structure, in which our aircraft could not operate;
• Presence of a 30-meters high power transmission tower just beside the reservoir, that not only created a physical obstacle to lower flights but also was a potential source of electromagnetic interference for the communication channel between our ground stations and the aircraft;
• Dense, high vegetation near both structures, creating additional obstacles to close flights.

Risk assessments were thoroughly reviewed with the client. Flight grid paths and our own no-fly zones were defined to guarantee a safe operation. Imagery capture spots were planned to ensure enough proper coverage of all areas of interest in the inspection and allow for the 3D reconstruction. To counterbalance all obstacles affecting proper aerial coverage, our ground team also made sure to complement the captured dataset with enough photos from the ground.

Back to the office, time for some classifying over the inspection imagery before shipping the dataset to the client.

This was also the moment for heating up our graphic servers’ processors and reconstruct realistic 3D models of the two scoped structures using only photos as inputs. For those who love to do this, fun time!

The resulting 3D models will allow engineers working for our client to plan maintenance and improvements without leaving their desks:

Want to integrate aerial asset inspections in your operation? Contact us today and we will be more than happy to help you get the most out of this revolution.

The identification of animals and objects in complex environments is an area of computer vision which has largely been neglected due to high expense of datasets and intensive computational requirements for analysis.  The challenges of identifying sharks and aquatic animals in environments littered with surfers, rocks and various other possible false positives required the development of a system incorporating the newest forms of computer vision machine learning. Simple methods such as color separation, intensity histograms and optical flow are all thrown off by the ever changing complex environment of the ocean.  For this reason a machine learning method that gradually improves with larger and larger datasets was the best solution.

The development of this system needs to overcome a few large obstacles that are present in this dynamic environment

• Ability to identify sharks from a minimum 60 meters altitude from a fast moving drone
• Detection through the highly changing environment of waves, glare and shadows created from the changing ocean
• False positives from surfers and a range of other marine life
• Ability to be run on an onboard single board computer

Examples of difficult environments

With use of machine learning methods the ability to essentially train out specific problematic areas and accentuate features of the shark for easier detection become feasible.  This has required the use of immensely huge datasets with over 50,000 individual images being utilized in order to successfully differentiate between the aggressive ocean and the targeted entity.  The result has been extremely successful over other techniques with low false positives and high identification in extremely complicated environments. The problems of waves, shadows and rocks have all been slowly trained out to a point where blind runs of unseen datasets produce high identification rates and low false positives.

The capabilities of this machine learning technique are near endless with the addition of more computational power the ability to run identification of multiple animals and objects can be undertaken. This can lead to a huge new range of possibilities in creating specific scenarios for warning systems and data collection.

Do you have a similar project in mind? Want to successfully integrate remotely piloted aircraft into your operation?  We would love to help you turn your idea into practical results. Contact us today.