Nonlinear tool traces fast tracing algorithm based on single point laser detection

Pan, Nan; Kan, Lifeng; Liu, Yi; Fu, Wei; Hou, Zhanwei; Li, Gang; Qian, Jun; Fu, Xiaodong

doi:10.3233/jifs-169885

Cited by 1 publication

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can also be used effectively in gun batch archiving and in identifying the guns used in a crime. However, in practical applications, the following problems still exist and need to be studied in depth: (17)(18)(19) 1) Although the detection accuracy of this system can meet the matching requirements, further research is needed on how to improve the detection and matching efficiency, such as transforming the signal into the transform domain, increasing the calculation speed, and further improving the matching accuracy. 2) In-depth research on the two-and three-dimensional fusion algorithms and the fusion of the data obtained by the laser sensor detection with the unfolded image data of the bullet surface can improve the matching accuracy.…”

Section: Discussionmentioning

confidence: 99%

“…The distance functions generally replaced by the similarity calculation include Euclidean distance, K-means, and so forth. (19) After many experiments, the Pearson correlation coefficient was finally selected to calculate the similarity in the trace feature data. We set the two trace detection data arrays as l 1 and l 2 and the Pearson correlation coefficient as r, which is defined as…”

Section: Similarity Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

Batch Filing System of Bullets’ Rifling Traces Based on Point Laser Detection

Xiang¹,

Pan²,

Pan³

et al. 2022

Sensors and Materials

Self Cite

View full text Add to dashboard Cite

The traces of bullets' rifling are not easy to destroy, difficult to disguise, and unique. It is of great significance for the batch filing of standard guns. Aiming at the problems of traditional binocular microscope comparison, segmented photo stitching, and other methods, namely, timeconsuming, low efficiency, and inability to carry out quantitative analysis, a batch filing system of bullets' rifling traces based on point laser detection was studied. Firstly, the translation, swing, and rotation of the test piece are controlled by the movement of five-stage stepper motors until the cylindrical centerline of the tested bullet coincides with the rotation centerline of the electric rotating table. Secondly, laser displacement sensing technology is used to detect rifling marks on the bullet surface in a 360-degree circle along the circumference direction, the superposition of multiple detection data is used to remove the influence item, and the generalized morphological filtering is used for the noise reduction of detection data. After that, the Pearson correlation coefficient is used to compare and calculate the similarity of traces, and finally achieve the rapid matching of bullet traces. With this system, a total of 321 bullets fired by 69 different guns were matched with rifling traces, and an excellent matching rate of more than 80% and a relatively high matching speed were obtained. The actual bullet matching experiments show that this system can be effectively applied to the batch filing of guns.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Similarity Comparisonmentioning

confidence: 99%