MEASUREMENT OF BLAST FRAGMENTATION USING IMAGE PROCESSING ANALYSIS 3D PHOTOGRAMMETRY.

Measurement of blast fragmentation using image processing (IP) analysis especially 2D method have been widely used because of its simplicity in aplied in the field. IP analysis method for measurement of rock fragment consists of 2D and 3D methode. In practice, the condition of rock fragment in muckp...

Full description

Saved in:
Bibliographic Details
Main Author: Hidayat, Syamsul
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/78505
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:Measurement of blast fragmentation using image processing (IP) analysis especially 2D method have been widely used because of its simplicity in aplied in the field. IP analysis method for measurement of rock fragment consists of 2D and 3D methode. In practice, the condition of rock fragment in muckpile which overlapping cause the part of fragment was not completely analyzed on 2D methods, and this condition can cause error estimation. This study used 3D photogrammetry technique for measurement of blast fragmentation. 3D photogrammetry technique was chosen because this technique able to model the hidden parts of fragment in muckpile that can not be appeared on the 2D method. This study was laboratory experiment and field observation activities. Measurement of rock fragment used IP analysis of 2D and 3D method in the laboratory test then applied to measure blast fragmentation in the field. Laboratory experiment was done to know the comparison between the size distribution of fragment that resulted from sieving (actual) and Image processing analysis (2D and 3D). Based on the result of laboratory experiment, the significance difference between sieving and 3D was found, so that, correction of fragment size distribution obtained from measurement used 3D photogrammetry technique was aplied to improve the distribution. The correction was aplied by using the linier regression equation and Monte Carlo random numbers methode. Error estimation can be reduced after aplied distribution correction for 3D photogrammetry result. Mean Absolute Percent Error (MAPE) between sieving and 2D analysis was 22.98%, sieving and 3D analysis (before corrected) was 27.1%. MAPE between sieving and corrected 3D analysis was 5.32%, it value was better than the result of sieving and 2D analysis. Field research consist of the observation of the rock mass condition, observation of blast design, measurement of blast fragmentation using IP analysis 2D and 3D method and prediction of blast fragmentation using Kuzram model, all of field research were performed at andesite quarry mining. The correction factor obtained from laboratory experiment was applied to the measurement of blast fragmentation in field muckpile. In field Research, MAPE between 2D and corrected 3D analysis was 29.85%, between Kuzram model and correected 3D analysis was 39.8%.