Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The overengineered solution to my pigeon problem (2022)

maxnagy.com

Post reply on HN
Print view View original

61–70 of 71 posts

Re: The overengineered solution to my pigeon problem (2022)

#61

Post by maxmunzel »

guest__user wrote: post #60
maxmunzel wrote: post #49

Hey, author here — thanks for posting it again! As of 2025 the pigeons are back and version 2 with aiming and a microphone array for target acquisition is still work-in-progress. I’m heading to bed now, but I’ll happily answer any questions tomorrow.

I am using a raspberry pi and a two axis turret with a pump and reservoir from a car windshield sprayer. right now it's just using differences in pixels to detect movement, by it it's keeping the balcony pigeon free! (a few false positives when a bumble bee flies too close to the camera though) I also have a yolo model version but haven't implemented it yet !

Cool! Do you have the STLs of the turret online somewhere?

My setup also had its occasional false positives but that’s not a problem if all you do is spray a bit of water around. I also looked into yolo, but I couldn’t find any good datasets of pigeons. Nowadays that’s no problem anymore as you can just use models like meta’s segment anything that do zero-shot bounding boxes… if you have the GPU to run it.

Re: The overengineered solution to my pigeon problem (2022)

#62

Post by maxmunzel »

Taek wrote: post #59
maxmunzel wrote: post #49

Hey, author here — thanks for posting it again! As of 2025 the pigeons are back and version 2 with aiming and a microphone array for target acquisition is still work-in-progress. I’m heading to bed now, but I’ll happily answer any questions tomorrow.

The introduction of GPT 4o probably gives you a much simpler path to identifying the location of pigeons right? Is that something you have looked into?

Not yet, but I used VLMs to read out water meters with mixed results. It is definitely the easiest approach to prototype. For my balcony, all vision based approaches are limited by viewing angles that don’t expose any neighbors.

But even with an llm backbone you’d still need some setup to detect when to query the llm. And then you already have a low-false-negative pigeon detector that may be sufficient for your use case.

Re: The overengineered solution to my pigeon problem (2022)

#63

Post by guest__user »

maxmunzel wrote: post #61
guest__user wrote: post #60

Earlier quoted context omitted.

I am using a raspberry pi and a two axis turret with a pump and reservoir from a car windshield sprayer. right now it's just using differences in pixels to detect movement, by it it's keeping the balcony pigeon free! (a few false positives when a bumble bee flies too close to the camera though) I also have a yolo model version but haven't implemented it yet !

Cool! Do you have the STLs of the turret online somewhere? My setup also had its occasional false positives but that’s not a problem if all you do is spray a bit of water around. I also looked into yolo, but I couldn’t find any good datasets of pigeons. Nowadays that’s no problem anymore as you can just use models like meta’s segment anything that do zero-shot bounding boxes… if you have the GPU to run it.

[dead]

Re: The overengineered solution to my pigeon problem (2022)

#64

Post by jijijijij »

jyounker wrote: post #31
jijijijij wrote: post #19

Earlier quoted context omitted.

I shoot them with a "pea shooter", a cardboard pipe with a balloon attached to one end, loaded with dry lentils or rice. Stretching/charging and releasing the balloon makes a loud, directed thump noise, but by itself it's quickly ignored by pigeons. They need to be actually hit by the load to care. (BTW I can confirm, although quite violent, pigeons seemingly don't readily learn from this experience... They are reall…

> They need to be actually hit by the load to care. (BTW I can confirm, > although quite violent, pigeons seemingly don't readily learn from > this experience... They are really fucking stupid.) That doesn't sound stupid to me. Rather it sounds like they're willing to put with a lot of annoyance in order to get food.

It's not about the food. It's about a place to rest their fat bodies. Lacking the agility of every other bird, they prefer strictly horizontal branches of a certain diameter. There is plenty of other trees around. They could just realize the particular ones in front of my window mean getting shot with stuff. Honestly, they could alternatively just contain themselves and not literally shit my plants kaputt. We could coexist.

Do you commonly see other birds disfigured because they kept ignoring deterrents? No?! That's because pigeons are fucking stupid.

Re: The overengineered solution to my pigeon problem (2022)

#65

Post by assemblyman »

righthand wrote: post #48
assemblyman wrote: post #42

As an aside, pigeons are deeply misunderstood. Here's a very nice article about it: https://www.bbc.com/future/article/20240620-why-do-people-pe... From the article: "One study found that, between 1941 and 2004, there were just 207 reports of pathogens transmitted from pigeons to humans – anywhere in the world. In all, there were 13 recorded deaths. The true number may be higher, but it would have to be off by severa…

Feeding pigeons is a bad thing. For one the pigeons are not starving and you are over feeding them. The other issue is that you are training them to depend up humans for food. You are also aiding in creating unnaturally large pigeon populations since they are overfed. Many places it is illegal to feed the birds. Media has created this stupid meme where people think it’s a hobby to over feed birds in cities.

I don't think feeding pigeons is objectively a bad thing. While I live in a highly urban area and don't doubt they get food, a couple of additional sessions of seeds per week is not going to make them dependent or obese. I generally confirm I am doing something reasonably appropriate by asking my local bird sanctuary.

I don't follow social media at all so not sure if there's a trend going on about feeding birds. A few times someone has shown me clearly fake (i.e. spliced together) videos of animals being "rescued" and living with families. They do make me angry because they are clearly made for views.

Re: The overengineered solution to my pigeon problem (2022)

#66

Post by ho_schi »

hypercube33 wrote: post #52
ho_schi wrote: post #40

Solution: Cut food supply through humans trash. Don’t allow people to drop food on the ground in city center or elsewhere. And make life for some natural predators comfortable, like crows. When I first saw how two crows shared a pigeon I was impressed.

Clearly you don't live near trains that carry grain. They leak corn all over and are basically a breeding ground for pigeons.

OMG. That sounds…not funny.

Re: The overengineered solution to my pigeon problem (2022)

#67

Post by zrn900 »

yegle wrote: post #38
Mikhail_Edoshin wrote: post #23

We have a pigeon on our balcony. We picked him when he was only a week old and now he lives with us. Our balcony is isolated from the street, so every morning we open a window there so Theodor (his name; although we're not sure, maybe he is she), so Theodor can go out and fly. In the evening he returns; sometimes he gets stuck somewhere due to rain (they do not fly in the rain) until dark (they do not fly in the dark…

> Pigeons are very pleasant to touch. The feathers have a silky feeling and they are warm, warmer than people. My first language is not English, so I learned this late: pigeons are perceived as dirty birds due to its presence in urban environments, but they are the same species of doves which is perceived as clean. In my first language they are just called "pigeon" and "white pigeon".

> pigeons are perceived as dirty birds due to its presence in urban environments, but they are the same species of doves which is perceived as clean

The delirious doublespeak of the English public discourse, that is. You dont see it in the rest of the world, ie, in the Mediterranean, even in Europe. There are gigantic plazas with flocks of pigeons that live there as a policy because people dont want to see plazas without pigeons.

For some reason, the English public discourse is hostile to living creatures - the same kind of 'flying pest' rhetoric is being applied to the colorful indian ringnecks that populated the ghastly, damp English public parks. Its like hating things is a cultural trait in the English public discourse, for some reason. Makes one understand where the English tabloid press that spreads hate to everyone comes from.

Re: The overengineered solution to my pigeon problem (2022)

#68

Post by zrn900 »

assemblyman wrote: post #42

As an aside, pigeons are deeply misunderstood. Here's a very nice article about it: https://www.bbc.com/future/article/20240620-why-do-people-pe... From the article: "One study found that, between 1941 and 2004, there were just 207 reports of pathogens transmitted from pigeons to humans – anywhere in the world. In all, there were 13 recorded deaths. The true number may be higher, but it would have to be off by severa…

> As an aside, pigeons are deeply misunderstood

Yep. To contract anything, you have to eat or lick the pigeon, eat or lick its droppings, and the pigeon needs to have salmonella etc, right at that moment.

Re: The overengineered solution to my pigeon problem (2022)

#69

Post by assemblyman »

zrn900 wrote: post #67
yegle wrote: post #38

Earlier quoted context omitted.

> Pigeons are very pleasant to touch. The feathers have a silky feeling and they are warm, warmer than people. My first language is not English, so I learned this late: pigeons are perceived as dirty birds due to its presence in urban environments, but they are the same species of doves which is perceived as clean. In my first language they are just called "pigeon" and "white pigeon".

> pigeons are perceived as dirty birds due to its presence in urban environments, but they are the same species of doves which is perceived as clean The delirious doublespeak of the English public discourse, that is. You dont see it in the rest of the world, ie, in the Mediterranean, even in Europe. There are gigantic plazas with flocks of pigeons that live there as a policy because people dont want to see plazas wit…

100% agree. In south asia and south east asia (maybe other parts too), it's very common to have insects like spiders, lizards, local bugs, ants inside apartments and houses. No one bats an eyelid and, if we felt, a larger insect was trapped and couldn't make it outside, we would just put them on paper tissue and take them outside.

In the US though, I frequently find people freaking out if they see any insect. There's a zero-tolerance policy for any living creature indoors. They are almost always killed. It leads to a disconnect between human beings and any other life. Other creatures are always a distant presence in zoos or on TV although exceptions are made for pets.

This shows up in language too. Instead of saying, a bear was killed for straying into settled human land or breaking into a house, the phrase used is "the bear was euthanized" (still accurate) or "the bear was destroyed" as if it was a piece of furniture. To contrast this with, say India, even tigers and elephants that kill, are mostly tranquilized and moved deep into a forest. This is very alien in the US where the trigger-happy reaction is to kill the animal.

Re: The overengineered solution to my pigeon problem (2022)

#70

Post by assemblyman »

vjerancrnjak wrote: post #45
ho_schi wrote: post #40

Solution: Cut food supply through humans trash. Don’t allow people to drop food on the ground in city center or elsewhere. And make life for some natural predators comfortable, like crows. When I first saw how two crows shared a pigeon I was impressed.

Let’s solve the problem with ingenuity, not brutality. Brutality is lazy. Dealing with animal overpopulation by introducing carnage. I know nature is indifferent but Homo sapiens should not be.

It's very heartening to read this sentiment. Thank you!
Post reply on HN