First, I want to simulate a small project with an indoor aquarium in the case of cloudy water and cannot use AI cameras to determine the number of fish, can anyone have any ideas on what types of sensors I can use? to perform this experiment?
Sounds like a very unrealistic project, difficult to impossible to implement, with a return of dubious value.
An Arduino is very unlikely to make any sort of contribution.
However, if you are determined to pursue it, start by ask your local fisheries officials about how they go about counting fish.
Accurately?
I've seen it done; it's a little hard on the fish.
And: Home Automation???
Thanks for your help, sorry for the confusion
Thanks , i will ask them for some solution about it
Empty the lake.
OR
Measure the pressure at the bottom of the lake.
Calculate the difference between "pure" water pressure and the measured pressure.
The difference will be the fish... and maybe some other stuff.
Most of whom are either in the act of being neutrally buoyant, or are resting on the bottom.
Besides, that just gives you the total mass of fish. If might be a million little ones or one great big one.
What you gotta do see is make a big fence as wide as the lake with one hole in the middle and drag it across from one side to the other so all the fishies have to swim through the hole at some point. Then you can put a break beam on the hole to count the fish.
Most of the fish could be up in the creeks spawning.
Catch half the fish and count them. Double that number.
2 Likes
Do something like this, but for fish.
Perfectly Counting Washers #shorts #short #trendingshorts#viralshorts
Drop a 5 ft wide pipe whose length is equal to the depth of the lake. Make sure you drop it fast so the fish can't dodge. Pump the water out of it. Count the fish. Multiply: fish_cnt * vol_lake / vol_pipe.
Lots of funny responses in this thread. Rightfully so since there really isn't much an Arduino can do for this project.
What I can tell you is that many years ago I worked a job that involved estimating the number of bald eagles or gopher tortoises or sand skinks in an area. The scientifically accepted method to estimate the total in a given area is to sample a smaller area and ultimately use the Monte Carlo method to approximate the total population. It's a little tricky because no arbitrary sample is going to be indicative of the whole, and so it takes a professional touch to come up with a decent "random" sample space that is roughly indicative of the whole. However, if we are talking fish in a tank and not in a natural area, it should be easy enough to make a fair sample.
Anyway, the basics of the Monte Carlo method is similar to throwing darts. You can estimate the size of the bulls eye by knowing the total size of the board, then throwing 1000 darts and counting how many landed in the bulls eye. That number of darts divided by the total number of darts is the same as the size of the bullseye divided by the size of the board. Hope this helps.
1 Like
If my darts performance is any indicator, then the bullseye must be infinitesimally small.
approaching infinitesimal
Anyway, I think I know what you're saying, but there really is a statistical method behind it. The wiki has a lot of good information on it, and I went the old fashioned way of going through Stat Theory I to get it. Either way, we agree that it's definitely the go-to method for this application where you are going to have a limited sensing volume, an idea of the total perimeter, and can 'throw 1000 darts' per second. On the other hand, Monte Carlo offers strictly an approximation which may or may not be enough for OP.
You could do it the way the fish and game commission do it in Oregon.
Idiot bass fishermen plant bass in lakes so they can catch them. In a while, the bass EAT all the other fish and soon only bass exist in the lake.
The solution is to block ALL input and output to the lake. Then poison the lake to kill ALL the fish, including bass. Then they gather all the dead fish. When all the bass are gone, trout and other natural fish are planted back into the lake.
You could count all the dead fish if you wanted. Perhaps the commission does that already.
That is an accurate a count as is possible.