100
2 Experiments in Pebble Flows
centers of the right end particle and the left end particle in the arch whose size is s
[56]. Besides, more details relating the algorithm to find bridges in packing and the
identification of arches can be obtained in the reference paper [53].
2.5.4.4 Improvement of RM and Verification
Before the detection of particles, the images are preprocessed to reduce background
noise, which includes convolving the image with a Gaussian and an average filter.
The circular-shaped filter has been added to the algorithm to improve the particle
tracking of circular-shape particles.
The synthetic particle images from Visualization Society of Japan (VSJ) [18]
are employed to test the DTB method. The VSJ supplies simulated particle image
sets for evaluation and testing of various preprocessing methods [12, 19, 20]. The
VSJ#302 set (Fig. 2.34) includes 147 frames with about 1000 particles in each frame.
The error analyses of particle location, measurement reliability, and RMS error are
Fig. 2.34 The standard image and validation of the DTB method. a The standard image from
Visualization Society of Japan (VSJ#302). b The percentage of detected particles within position
errors. c The corresponding reliability and errors
2 Experiments in Pebble Flows
centers of the right end particle and the left end particle in the arch whose size is s
[56]. Besides, more details relating the algorithm to find bridges in packing and the
identification of arches can be obtained in the reference paper [53].
2.5.4.4 Improvement of RM and Verification
Before the detection of particles, the images are preprocessed to reduce background
noise, which includes convolving the image with a Gaussian and an average filter.
The circular-shaped filter has been added to the algorithm to improve the particle
tracking of circular-shape particles.
The synthetic particle images from Visualization Society of Japan (VSJ) [18]
are employed to test the DTB method. The VSJ supplies simulated particle image
sets for evaluation and testing of various preprocessing methods [12, 19, 20]. The
VSJ#302 set (Fig. 2.34) includes 147 frames with about 1000 particles in each frame.
The error analyses of particle location, measurement reliability, and RMS error are
Fig. 2.34 The standard image and validation of the DTB method. a The standard image from
Visualization Society of Japan (VSJ#302). b The percentage of detected particles within position
errors. c The corresponding reliability and errors