Detecting Arrowheads with drones & computer vision

[pkrush]

It’s getting way easier to find stuff with computer software. For example you can train a program to understand the difference between a rock and an arrowhead in an image from a camera. I still think a human would be better unless camera was able to travel more ground, or “ground” was moving past the camera on a conveyor belt. Drones are really cool, but I have a feeling setting up lines for the camera to travel cable car style would be the way to go. Open ground and lots of light would be best.

Of course one could look for whatever they can see with their eyes: minerals, gems, fossils, and other relics. There would be an advantage in looking for tiny stuff. It’s easier to make the camera cover more ground and it could see small things much quicker than your eyes can.

What do you think? Do you see this happening? I do, it’s not half baked sci-fi. It wouldn’t be perfect, but it would speed up the hunt. I would love to build the equipment to do this. I am working with coins right now, but this tech (deep learning) is easy to adapt.

Out of curiosity what sorts of things would you look for? What area do you think people would be the most interested? Scanning for their lost keys?

Thanks!

Paul Krush
GemHunt.com

 

Amazon Forum Fav 👍

A Handbook of Civil War Bullets and Cartridges - Grab it through Amazon!

[Charlie P. (NY)]

Certainly possible. My foolish CoolPix camera has software that doesn't snap a group image until every face in the frame has both eyes open and is smiling. It looks like a missile targeting display if you watch it "think" on the screen. It wasn't even a feature I knew it had until I started playing around some time after owning it. I forgot it was on and I couldn't figure why it wasn't letting me take a picture of my cat. I guess he wasn't grinning at the time.

But the problem is it will only work if the arrowhead is lying flat and fully exposed. Partials may not trigger it. You also need pretty good resolution.. In this area there is a LOT of shale that has similar shape just lying around.

And the next problem: it won't be cheap ($200 for the drone, maybe $150 for the camera and software) plus transmitter. And the joy of arrowheads is the satisfaction of finding one. If you start popping them out with a harvester they'll lose a lot of attraction.

 

[FreeBirdTim]

Can the drone pick up the arrowhead for you and bring it back? If not, how are you going to find the arrowhead? I don't see the point of a drone spotting arrowheads or anything else. You're going to have to go and retrieve the item yourself anyway, so what is gained?

Bottom line, if you're so lazy that you need a drone to find arrowheads for you, maybe you should take up a different hobby...

 

[TheHunterGT]

Can the drone pick up the arrowhead for you and bring it back? If not, how are you going to find the arrowhead? I don't see the point of a drone spotting arrowheads or anything else. You're going to have to go and retrieve the item yourself anyway, so what is gained?

Bottom line, if you're so lazy that you need a drone to find arrowheads for you, maybe you should take up a different hobby...

Just because he is thinking outside of the box does not mean he is lazy and should pick up another hobby. Obviously the drone finds the object...then the human goes to retrieve it once found. Not very complicated what he is talking about. The gain would be time since drone does the time-consuming work which is the hunting aspect. Imagine if you went to the local park and all the coins had flags sticking out for you to go dig without swinging a detector. Time would be saved/gained.

I do think we are many years from anything like this however. Consumer software that could accomplish this would be very pricey. The camera it would take to get such high-resolution video footage would also be cost prohibitive for most. A $500 Go-Pro would not come close. also the cost of the drone that could have such an extended range and flight time would be very costly.

Beyond cost and technology constraints.....best case scenario I could see would be a drone running a grid pattern of an area then fly to an extraction point once finished. After pick-up the software checks the footage for objects it deems worthy for investigation and marks them on the grid via a printed map or whatever. User takes map and goes on foot to check each marker.

Like I said a ways off...but I'd be game if the costs were not crazy high.

 

[Charlie P. (NY)]

I don't think you'd want a GoPro. You'd want real time first-person view feed or recognition software fed back to a PC monitor. No sense recording hours of overflying corn stalks and dirt. If you're running it through recognition software you might not want to see anything until it alerts you to a potential "hit". You'd probably have to use edge-detection alogrithms and flag only the objects that matched an arrowhead/spearhead shape.

 

[TheHunterGT]

Real-time would be cool. But software/hardware that would handle real-time computations like that over slowly processing footage would be even more cost prohibitive to the consumer.

Not so sure I would want to drag the drone plus hardware (monitor/computer) it would take to do the processing out in the field.

I would want a drone to lift off from my back yard....go to designated way-points and then return for me to analyze. Then decide if each hit was a viable choice. How's that for lazy!

 

[BosnMate]

I found the very best way to hunt arrowheads is after a rain, on a saddle horse. The horse watches where it's going, all your attention is looking, the horse covers a lot of ground, and the points just shine up at you.

This is the first point I found horse back, since then there have been many others.

 

[T.C.]

Bee-u-ti-ful jasper Rose Spring!!

 

[GaRebel1861]

Think of the other items that could and would get missed that are not shaped nowhere near an arrowhead or spear. Bannerstones, celts, beads, etc. Cool idea though. Kind of like when the discriminator was added on the detectors so we didn't have to dig pull tabs any longer. It was a good idea but, I still ended up digging pull tabs so I don't miss a ring.