Psych 221 Image Systems Engineering - User contributions [en]

ISETHDR CV Experiment

2025-12-08T19:39:35Z

Lschul: /* Image generation */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

::{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

::{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1114031438_ar0132at_day_07_8.0ms.png|200px]]
| [[File:id1114031438_ar0132at_dusk_07_8.0ms.png|200px]]
| [[File:id1114031438_ar0132at_night_07_8.0ms.png|200px]]
| [[File:id1114031438_ar0132at_blind_07_8.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | Scene 1114031438 under day, dusk, night, and blind light scenarios with an exposure time of 8 ms
|}

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

Dynamic range, field of view, and resolution are some sensor parameters that are relevant in driving applications. In this project, we examine the influence of dynamic range on object identification. A large field of view is advantageous; since the car needs to be able to sense 360 degrees around its surrounding, larger field of view means that fewer sensors are required. A higher resolution, however, is not necessarily advantageous. The objective is to simply identify objects, rather than distinguish fine details, so a high resolution may require excessive data transmission and processing. We consider the field of view and resolution (table below) but do not vary them in the experiment.

::{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

::''Field of view is calculated for a 4 mm focal length''

In order to vary dynamic range, we sweep exposure time across 14 values from 0.1 ms to 1000 ms. In all, we sweep 4x scenes, 4x light scenarios, 4x sensors, and 14x exposure times, for a total of 896 images.

=== YOLO ===

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

File:Id1114031438 ar0132at night 07 8.0ms.png

2025-12-08T19:33:06Z

Lschul:

File:Id1114031438 ar0132at dusk 07 8.0ms.png

2025-12-08T19:32:53Z

Lschul:

File:Id1114031438 ar0132at day 07 8.0ms.png

2025-12-08T19:32:40Z

Lschul:

File:Id1114031438 ar0132at blind 07 8.0ms.png

2025-12-08T19:32:22Z

Lschul:

ISETHDR CV Experiment

2025-12-08T19:29:04Z

Lschul: /* Methods */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

::{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

Dynamic range, field of view, and resolution are some sensor parameters that are relevant in driving applications. In this project, we examine the influence of dynamic range on object identification. A large field of view is advantageous; since the car needs to be able to sense 360 degrees around its surrounding, larger field of view means that fewer sensors are required. A higher resolution, however, is not necessarily advantageous. The objective is to simply identify objects, rather than distinguish fine details, so a high resolution may require excessive data transmission and processing. We consider the field of view and resolution (table below) but do not vary them in the experiment.

::{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

::''Field of view is calculated for a 4 mm focal length''

In order to vary dynamic range, we sweep exposure time across 14 values from 0.1 ms to 1000 ms. In all, we sweep 4x scenes, 4x light scenarios, 4x sensors, and 14x exposure times, for a total of 896 images.

=== YOLO ===

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

ISETHDR CV Experiment

2025-12-08T19:28:38Z

Lschul: /* Image generation */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

::{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

:: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

Dynamic range, field of view, and resolution are some sensor parameters that are relevant in driving applications. In this project, we examine the influence of dynamic range on object identification. A large field of view is advantageous; since the car needs to be able to sense 360 degrees around its surrounding, larger field of view means that fewer sensors are required. A higher resolution, however, is not necessarily advantageous. The objective is to simply identify objects, rather than distinguish fine details, so a high resolution may require excessive data transmission and processing. We consider the field of view and resolution (table below) but do not vary them in the experiment.

::{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

::''Field of view is calculated for a 4 mm focal length''

In order to vary dynamic range, we sweep exposure time across 14 values from 0.1 ms to 1000 ms. In all, we sweep 4x scenes, 4x light scenarios, 4x sensors, and 14x exposure times, for a total of 896 images.

=== YOLO ===

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

ISETHDR CV Experiment

2025-12-08T19:27:59Z

Lschul: /* Image generation */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

Dynamic range, field of view, and resolution are some sensor parameters that are relevant in driving applications. In this project, we examine the influence of dynamic range on object identification. A large field of view is advantageous; since the car needs to be able to sense 360 degrees around its surrounding, larger field of view means that fewer sensors are required. A higher resolution, however, is not necessarily advantageous. The objective is to simply identify objects, rather than distinguish fine details, so a high resolution may require excessive data transmission and processing. We consider the field of view and resolution (table below) but do not vary them in the experiment.

::{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

::''Field of view is calculated for a 4 mm focal length''

In order to vary dynamic range, we sweep exposure time across 14 values from 0.1 ms to 1000 ms. In all, we sweep 4x scenes, 4x light scenarios, 4x sensors, and 14x exposure times, for a total of 896 images.

=== YOLO ===

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

ISETHDR CV Experiment

2025-12-08T19:26:10Z

Lschul: /* Image generation */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

Dynamic range, field of view, and resolution are some sensor parameters that are relevant in driving applications. In this project, we examine the influence of dynamic range on object identification. A large field of view is advantageous; since the car needs to be able to sense 360 degrees around its surrounding, larger field of view means that fewer sensors are required. A higher resolution, however, is not necessarily advantageous. The objective is to simply identify objects, rather than distinguish fine details, so a high resolution may require excessive data transmission and processing. We consider the field of view and resolution (table below) but do not vary them in the experiment.

::{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

::''Field of view is calculated for a 4 mm focal length''

YOLO

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

ISETHDR CV Experiment

2025-12-08T19:11:44Z

Lschul: /* Image generation */

== Introduction ==

== Background ==

== Methods ==

=== Image generation ===

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

{| class="wikitable" style="border:none; text-align:center;"
|-
| [[File:id1112201236_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1112184733_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1113094429_ar0132at_day_05_4.0ms.png|200px]]
| [[File:id1114031438_ar0132at_day_05_4.0ms.png|200px]]
|-
| colspan="4" style="text-align:center" | ISET HDR Scenes 1112201236, 1112184733, 1113094429, and 1114031438
|}

To set up the scenes, we consider four lighting scenarios that commonly occur during driving. The light scenarios are defined by a vector of weights for headlights, streetlights, other lights, sky map, in order. The daytime scenario has strong illumination from the sky only. The nighttime scenario is illuminated almost only by headlights and streetlights. The dusk scenario falls between day and night; it has half of the daytime sky illumination combined with headlights and streetlights. Finally, the blind scenario represents a nighttime scenario with stronger artificial lighting; the headlights and streetlights are 10x greater than in the nighttime scenario.

: ''Illumination vector: [ headlights, streetlights, other lights, sky map ]''
* Day - [ 0, 0, 0, 50 ]

* Night - [ 0.2, 0.001, 0, 0.0005 ]

* Dusk - [ 0.2, 0.001, 0, 20 ]

* Blind - [ 2, 0.1, 0, 0.0005 ]

Next, we need to set up the camera. We selected four sensors. The ar0123at and mt9v024 are single pixel sensors for automotive applications while the ov2312 is a split pixel sensor also for automotive applications. In contrast, the imx363 is a single pixel sensor for smartphone applications.

{| class="wikitable"
! style="text-align:left; padding: 0 12px;"| Sensor
! style="padding: 0 12px;"| Pixel type
! style="text-align:center; padding: 0 12px;"| Resolution
! style="text-align:center; padding: 0 12px;"| Dynamic range
! style="padding: 0 12px;"| Application
! style="text-align:center; padding: 0 12px;"| FOV
|-
| style="padding: 0 12px;"| AR0132AT
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 1.2 MP
| style="text-align:center; padding: 0 12px;"| 115 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 76°
|-
| style="padding: 0 12px;"| MT9V024
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 0.4 MP
| style="text-align:center; padding: 0 12px;"| 100 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 69°
|-
| style="padding: 0 12px;"| IMX363
| style="padding: 0 12px;"| Single
| style="text-align:center; padding: 0 12px;"| 12 MP
| style="text-align:center; padding: 0 12px;"| n/a
| style="padding: 0 12px;"| Smartphone
| style="text-align:center; padding: 0 12px;"| 21°
|-
| style="padding: 0 12px;"| OV2312
| style="padding: 0 12px;"| Split
| style="text-align:center; padding: 0 12px;"| 2 MP
| style="text-align:center; padding: 0 12px;"| 68 dB
| style="padding: 0 12px;"| Automotive
| style="text-align:center; padding: 0 12px;"| 81°
|}

YOLO

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]

ISETHDR CV Experiment

2025-12-08T17:31:36Z

Lschul: /* Methods */

File:Id1112201236 ar0132at day 05 4.0ms.png

2025-12-08T17:06:19Z

Lschul:

File:Id1112184733 ar0132at day 05 4.0ms.png

2025-12-08T17:05:53Z

Lschul:

File:Id1113094429 ar0132at day 05 4.0ms.png

2025-12-08T17:02:10Z

Lschul:

File:Id1114031438 ar0132at day 05 4.0ms.png

2025-12-08T16:59:45Z

Lschul:

ISETHDR CV Experiment

2025-12-08T16:55:51Z

Lschul: /* Methods */

== Introduction ==

== Background ==

== Methods ==

Image generation

First, we need to assemble a dataset of driving images to run the YOLO algorithm on. We acquired four driving scenes from the ISR HDR Sensor Repository. Each scene includes .exr files that contain radiance data for the sky, street lights, headlights, and other lights.

Scene images <insert images>
1112184733
1112201236
1114031438
1113094429

When setting up the scenes, we consider four lighting scenarios that commonly occur while driving. The light scenarios are described by a weight for each .exr file.

<insert table>
Day - strong sky illumination
[0, 0, 0, 50]

Dusk - weak sky illumination with headlights and streetlights
[0.2, 0.001, 0, 20]

Night - headlights and streetlights
[0.2, 0.001, 0, 0.0005]

Blind - extreme headlights and streetlights
[2, 0.1, 0, 0.0005]

Next, we set up the camera. We select four sensors:

That sets up the scenes. Now we set up the cameras.

We use 4 different sensor types: three automotive and one smartphone.

Automotive sensors

ar0123at: 1.2 MP, dynamic range >115 dB
mt9v024: 0.36 MP, dynamic range >100 dB
ov2312 split pixel: 2 MP, dynamic range >68 dB

Smartphone sensor
imx363: 12 MP

Now build the data set

use exposure time to alter dynamic range

times = 0.1, 0.5, 1, 2, 4, 6, 8, 12, 16, 20, 50, 100, 500, 1000 ms

Loop over:
4x scenes
4x lighting scenarios
4x sensors
14x exposures

total images - 896

YOLO

== Results ==

== Conclusions ==

== Appendix ==

[[File:Snip 20210106183207.png|200px]]