- Jul 23, 2023
- 115
- 70
What is everyone doing for their LPR capture software? Right now, I'm using Blue Iris with platerecognizer.com as the Plate Recognizer for ALPR, but it has some limits.
Namely, it only seems to capture 1 snapshot from an approaching vehicle, so you have to activate it at just the right time with zones. (unless I'm totally missing something - which may be because I'm pretty new to this). But because it only seems to capture 1 screenshot, in my scenario it makes it pretty tough to arrange zones to capture at just the right time because I need to capture in both directions up and down my street. (Where I'm from, cars are only required to have a plate on EITHER the front OR back - so I can't just capture half the lane, because on the same view I might need to capture the front -or- the back plate depending on which way the vehicle is traveling). I have 2 camera's set up for this, right next to the road, one pointed left, down the road, the other right, up the road. To give an idea of proximity, the camera's are mounted on my mailbox. I'm getting GREAT video daytime and night, the license plates are very clear in the video, ie if I play through a clip and pause it when the LP is in good view... Meaning... that I can manually track a plate down pretty easily... but doing it in automation has proven to be a bit difficult... so I'm thinking that maybe there is a better solution than what I'm currently using for this purpose, so I'm curious, what are you guys using?
Here's my set up. I have set up the left camera with 5 zones, A, B, C, D, E
I have my triggers set up such that A>B,C>D,E - Meaning it triggers if motion detected:
1) In zone B if there has been motion previously in zone A
2) In zone D if there has been motion previously in zone C
3) Any motion detected in zone E
Here's my set up. I have set up the left camera with 3 zones, A, B, C
I have my triggers set up such that A>B,C>D - Meaning it triggers if motion detected:
1) In zone B if there has been motion previously in zone A
2) In zone D if there has been motion previously in zone C
Layout:
This works FAIRLY well... here are some captures:
Left - Front
Left - Rear
Right - Front
Right - Rear
But then I also get some completely missed images, mostly from the right side, or just barely on - so out of focus or blurry, ie:
Whereas if I move this vehicle back a bit manually in the recorded video clip, I get a great image of the plate.
Or here... left front bad
So... I'm wondering if there is anything better (than platerecognizer.com) out there that either A) captures multiple snapshots and "tries" to find the best plate, or uses the video clip to render a plate?
-OR- if you have any other suggestions?
Thanks!!!
Namely, it only seems to capture 1 snapshot from an approaching vehicle, so you have to activate it at just the right time with zones. (unless I'm totally missing something - which may be because I'm pretty new to this). But because it only seems to capture 1 screenshot, in my scenario it makes it pretty tough to arrange zones to capture at just the right time because I need to capture in both directions up and down my street. (Where I'm from, cars are only required to have a plate on EITHER the front OR back - so I can't just capture half the lane, because on the same view I might need to capture the front -or- the back plate depending on which way the vehicle is traveling). I have 2 camera's set up for this, right next to the road, one pointed left, down the road, the other right, up the road. To give an idea of proximity, the camera's are mounted on my mailbox. I'm getting GREAT video daytime and night, the license plates are very clear in the video, ie if I play through a clip and pause it when the LP is in good view... Meaning... that I can manually track a plate down pretty easily... but doing it in automation has proven to be a bit difficult... so I'm thinking that maybe there is a better solution than what I'm currently using for this purpose, so I'm curious, what are you guys using?
Here's my set up. I have set up the left camera with 5 zones, A, B, C, D, E
I have my triggers set up such that A>B,C>D,E - Meaning it triggers if motion detected:
1) In zone B if there has been motion previously in zone A
2) In zone D if there has been motion previously in zone C
3) Any motion detected in zone E
Here's my set up. I have set up the left camera with 3 zones, A, B, C
I have my triggers set up such that A>B,C>D - Meaning it triggers if motion detected:
1) In zone B if there has been motion previously in zone A
2) In zone D if there has been motion previously in zone C
Layout:
This works FAIRLY well... here are some captures:
Left - Front
Left - Rear
Right - Front
Right - Rear
But then I also get some completely missed images, mostly from the right side, or just barely on - so out of focus or blurry, ie:
Whereas if I move this vehicle back a bit manually in the recorded video clip, I get a great image of the plate.
Or here... left front bad
So... I'm wondering if there is anything better (than platerecognizer.com) out there that either A) captures multiple snapshots and "tries" to find the best plate, or uses the video clip to render a plate?
-OR- if you have any other suggestions?
Thanks!!!