# How to Troubleshoot Common Issues With License Plate Recognition Cameras

By VDS Editorial · Published November 2026

Source: https://visiondetectionsystems.com/blog/lpr-camera-troubleshooting

> When a license plate recognition camera misses plates, the cause is usually one of a handful of things. This guide matches each symptom to its likely cause and the fix to try first.

To troubleshoot a license plate recognition camera, start with a still frame of a vehicle it missed. That one image tells you which of four things went wrong: the plate was blurred, the plate was too small or too angled, the plate was over- or underexposed, or the plate was captured cleanly and the problem is further down the line in the software. Each has a different fix, and guessing between them wastes time.

The sections below go symptom by symptom. For the full list of recommended values, keep our [LPR camera settings guide](/blog/lpr-camera-settings-guide) open alongside.

## Symptom-to-cause quick reference

| What you see | Most likely cause | First thing to try |
|---|---|---|
| Plate is a streak or smear | Shutter too slow | Set a fixed fast shutter |
| Plate is a white rectangle at night | Too much infrared, or gain too high | Lower infrared intensity, cap gain |
| Plate lost next to headlights | Glare | Enable highlight compensation, adjust angle |
| Works by day, fails at night | Automatic exposure slowing down | Lock shutter, force night mode on a schedule |
| Wrong characters | Too few pixels, steep angle | Tighten field of view, straighten the camera |
| Some vehicles never appear | Capture zone, trigger or frame rate | Check where and when the camera captures |
| Reads fine, no alert or gate action | List or rule configuration | Check the plate format and list entries |
| Got worse over time | Dirty lens, shifted mount, new obstruction | Clean, re-aim, review the scene |

## Plates are blurry

First work out whether the blur is from motion or from focus. Pause on a frame with a moving vehicle and a stationary object at the same distance, such as a bollard or gate post.

- **Vehicle blurred, post sharp:** motion blur. The shutter is too slow. Cameras on automatic exposure lengthen the shutter as light fades, so this often shows up only in the evening. Set a fixed shutter; 1/1000 of a second or faster is a common starting point for moving vehicles.
- **Both soft:** focus. Refocus manually at the point where plates are read and lock it. If the image is sharp by day and soft at night, refocus under infrared, because some lenses shift focus between visible and infrared light.
- **Blocky or smeared detail:** compression or noise reduction. Raise the bitrate on the plate stream and turn digital noise reduction down.

## Reads fail at night

Night is where most plate cameras fall down, and there are three usual culprits.

**The plate is washed out.** Plates are reflective, so infrared that is too strong for the distance turns the plate into a blank white shape. Lower the infrared intensity or cap the gain until the characters are dark against the plate. A correct night image is mostly black with a clear plate in it.

**The plate is too dark.** The illuminator may not reach the capture point, may be aimed elsewhere, or may be blocked. Check that the infrared is on at night and that its range suits the distance.

**The camera keeps switching modes.** Headlights can push a camera on automatic day/night switching back into day mode at the wrong moment. Put night mode on a schedule.

## Headlight and sun glare

Headlights sit right beside the front plate, and a camera exposing for the whole scene lets them overwhelm it. Highlight compensation, where the camera offers it, masks the brightest areas. A slightly higher or more offset mounting position that takes the lens out of the direct beam helps too, as long as the angle to the plate stays shallow. Reading rear plates instead of front plates avoids headlights altogether where the layout allows it.

Low sun causes a similar problem for an hour or so at dawn or dusk. If reads drop at the same time each day, look at where the sun is. A lens hood, a small change in aim, or testing WDR at that time of day are the options.

## Wrong characters or partial reads

When the software returns a plate with one or two characters wrong, the image usually has too little detail for it to tell similar shapes apart.

- **Plate too small.** Check the character height in pixels on a still frame. A common rule of thumb is 20 to 30 pixels. If you are below that, zoom in or move the camera closer.
- **Angle too steep.** Characters viewed from the side or from above get squeezed together. Aim to stay within about 15 to 30 degrees of straight-on.
- **Plate obscured.** Frames, tow hitches, bike racks, dirt and damage hide parts of characters. No setting fixes these; an overview camera that records the vehicle gives you something to fall back on.
- **Unfamiliar plate designs.** Specialty plates, stacked characters and out-of-state formats are harder for any recognition software. Confirm that the software is set to the right region.

## Vehicles are missed completely

If some vehicles produce no read at all, the plate may never have been in the right place at the right time.

- **Capture zone.** Vehicles may be passing outside the area the camera covers, such as cutting a corner or using a second lane. Watch recorded footage of a missed vehicle to see its path.
- **Trigger settings.** If reads depend on motion detection or a detection zone, check that the zone covers the lane and that sensitivity is high enough at night.
- **Speed and frame rate.** A fast vehicle may be in the capture zone for only a few frames. Raise the frame rate, or slow traffic with a stop line or speed bump.
- **Tailgating.** A vehicle following closely can hide the plate of the one in front, or have its own plate hidden. Lane design is the fix here, not the camera.

## The read is right but nothing happens

Sometimes the camera is doing its job and the breakdown is in what happens next. If the plate shows up correctly in the log and no alert fires or the gate stays closed, check:

- that the plate is entered on the list in the same format the software outputs, including spaces, dashes and state;
- that similar characters were not typed incorrectly when the list was built;
- that the rule tied to the list is enabled for that camera and that time of day;
- that the camera has a working network connection and its clock is correct, since delayed or mis-timed reads can look like missed ones.

This matching step is handled by the [license plate recognition analytics](/platform/analytics/license-plate-recognition) layer. Testing it with a known vehicle before going live saves a lot of confusion later.

## Performance has dropped over time

A camera that used to read well and has slowly got worse has usually had something physical change.

- **Lens.** Dust, water spots, spider webs and road film all cut contrast, and webs light up under infrared. Clean the lens and housing window.
- **Mount.** Wind, vibration or a knock can shift the aim by a few degrees. Compare the current view with a saved reference image.
- **Scene.** New signage, parked equipment, vegetation or a changed traffic pattern can block or move the capture point.
- **Lighting.** A new floodlight or a failed one changes the exposure the camera was tuned for.

A camera on a [mobile surveillance trailer](/products/mobile-surveillance-unit) needs this check each time the unit is moved, because distance and angle change with every placement.

## Weather

Rain, snow and fog reduce reads on any camera, and no setting removes that completely. What helps: a hood or housing that keeps water off the front window, a shorter capture distance so there is less weather between the lens and the plate, and infrared positioned so it does not light up the rain directly in front of the lens.

## When to stop tuning

If you have a fixed fast shutter, controlled infrared, a tight single-lane view and a shallow angle, and reads are still unreliable, the hardware or its position is probably not suited to the lane. A general-purpose security camera can read plates in easy conditions and will struggle at speed, at distance and at night. At that point it is worth looking at a purpose-built unit; our [license plate recognition camera guide](/blog/license-plate-recognition-camera-guide) covers what to look for, and [VDS LPR](/products/lpr) is built for plate capture at sites without existing infrastructure. You can also [talk to our team](/get-started) about the specific entrance.

## Frequently asked questions

### Why is my license plate recognition camera not reading plates?

The most common causes are motion blur from a slow shutter, a plate that is too small in the frame, a steep camera angle, and overexposure from infrared or headlights at night. Check a still frame of a missed vehicle to see which one applies.

### Why are plates blurry on my LPR camera?

Blur on a moving plate almost always means the shutter is too slow, often because exposure is set to automatic and lengthens in low light. Blur on a stationary plate points to focus, a dirty lens or heavy compression.

### Why does my LPR camera work in the day but not at night?

At night the camera is either lengthening its exposure, raising gain, or flooding the reflective plate with infrared. Lock a fast shutter, cap the gain, and adjust infrared intensity until the characters are dark against the plate.

### Why does my LPR camera read the wrong characters?

Misreads of similar characters such as 0 and O or 8 and B usually come from too few pixels on the plate, a steep angle, or a plate frame covering part of the characters. Tighten the view and straighten the angle before changing software settings.

### How often should an LPR camera be checked?

Check it after any change to the site or the camera position, after severe weather, and on a regular schedule that includes cleaning the lens and reviewing a sample of night reads.