Research Article

Observation Selection, Total Variation, and L-Curve Methods for LiDAR Data Denoising

Table 1

The comparison of noisy observations removal by the “Newsom” method, “Baranov” method, HKO method, and proposed method.

RangeDataSNR SNR SNR

359 m–3509 m (total number: 1036067)LiDAR observations103326810328851031484
Denoising values by the “Newsom” method103147210314721031472
Denoising values by the “Baranov” method103326810328851031484
Denoising values by the HKO method103321210328741031484
Denosing values by the proposed method103298710327931031394
3509 m–6659 m (total number: 836510)LiDAR observations793282772883678138
Denoising values by the “Newsom” method677112677112677112
Denoising values by the “Baranov” method793282772883678138
Denoising values by the HKO method791440772670678130
Denosing values by the proposed method786772772632678043
6659 m–9809 m (total number: 455697)LiDAR observations28192718442577073
Denoising values by the “Newsom” method766457664576645
Denoising values by the “Baranov” method28192718442577073
Denoising values by the HKO method26303318302677018
Denoising values by the proposed method24771318435377070