Fig. 6
From: A hybrid zero-reference and dehazing network for joint low-light underground image enhancement
Comparison of safety helmet detection results before and after subsurface image enhancement. The blue box shows that the detection confidence of the safety helmet detection model after enhancement is significantly improved, and the yellow box shows the accurate detection of the enhanced image after missing the original image. The result proves that downhole image enhancement can enhance the feature of the target object, which is conducive to the monitoring of downhole vision equipment.
