A self-adaptive DBSCAN-based method for wafer bin map defect pattern classification

Chen, Shouhong; Yi, Mulan; Zhang, Yuxuan; Hou, Xingna; Shang, Yuling; Yang, Ping

doi:10.1016/j.microrel.2021.114183

Cited by 6 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The improved Harmony search algorithm (HS) was also used to optimize DBSCAN, when KDBSCAN combined with HS to optimize the clustering parameters Eps and MinPts [20]. Chin et al [21], proposed the use of DBSCAN as a Self-Adaptive DBSCAN-based method for solving a problem with wafer bin map such as the noise point detection of wafer maps (DBSCANWBM).…”

Section: Processmentioning

confidence: 99%

Intelligent Bat Algorithm for Finding Eps Parameter of DbScan Clustering Algorithm

Ghathwan

Mohammed²

2022

Iraqi Journal of Science

View full text Add to dashboard Cite

Clustering is an unsupervised learning method that classified data according to similarity probabilities. DBScan as a high-quality algorithm has been introduced for clustering spatial data due to its ability to remove noise (outlier) and constructing arbitrarily shapes. However, it has a problem in determining a suitable value of Eps parameter. This paper proposes a new clustering method, termed as DBScanBAT, that it optimizes DBScan algorithm by BAT algorithm. The proposed method automatically sets the DBScan parameters (Eps) and finds the optimal value for it. The results of the proposed DBScanBAT automatically generates near original number of clusters better than DBScanPSO and original DBScan. Furthermore, the proposed method has the ability to generate high quality clusters with minimum entropy [ 0.2752, 0.4291] in TR11 and TR12 datasets.

show abstract

Section: Processmentioning

confidence: 99%

Intelligent Bat Algorithm for Finding Eps Parameter of DbScan Clustering Algorithm

Ghathwan

Mohammed²

2022

Iraqi Journal of Science

View full text Add to dashboard Cite

show abstract

“…This method can avoid manual intervention and realize adaptive parameter optimization in the clustering process; however, the algorithm also needs to manually adjust the size of minPts. In [44], a new algorithm, SA-DBSCANWBM, was proposed. This method selects a comprehensive index of clustered intra-cluster density and inter-cluster density to evaluate the optimal parameters.…”

Section: Introductionmentioning

confidence: 99%

Path Planning Algorithm Based on Obstacle Clustering Analysis and Graph Search

Wang

Sun

2023

Symmetry

View full text Add to dashboard Cite

Path planning is receiving considerable interest in mobile robot research; however, a large number of redundant nodes are typically encountered in the path search process for large-scale maps, resulting in decreased algorithmic efficiency. To address this problem, this paper proposes a graph search path planning algorithm that is based on map preprocessing for creating a weighted graph in the map, thus obtaining a structured search framework. In addition, the reductions in the DBSCAN algorithm were analyzed. Subsequently, the optimal combination of the minPts and Eps required to achieve an efficient and accurate clustering of obstacle communities was determined. The effective edge points were then found by performing obstacle collision detection between special grid nodes. A straight-line connection or A* planning strategy was used between the effective edge points to establish a weighted, undirected graph that contained the start and end points, thereby achieving a structured search framework. This approach reduces the impact of map scale on the time cost of the algorithm and improves the efficiency of path planning. The results of the simulation experiments indicate that the number of nodes to be calculated in the search process of the weighted graph decreases significantly when using the proposed algorithm, thus improving the path planning efficiency. The proposed algorithm offers excellent performance for large-scale maps with few obstacles.

show abstract

A Defect Detection Method of Mixed Wafer Map Using Neighborhood Path Filtering Clustering Algorithm

Hou,

Qin,

et al. 2024

J Electron Test

View full text Add to dashboard Cite

A self-adaptive DBSCAN-based method for wafer bin map defect pattern classification

Cited by 6 publications

References 16 publications

Intelligent Bat Algorithm for Finding Eps Parameter of DbScan Clustering Algorithm

Intelligent Bat Algorithm for Finding Eps Parameter of DbScan Clustering Algorithm

Path Planning Algorithm Based on Obstacle Clustering Analysis and Graph Search

A Defect Detection Method of Mixed Wafer Map Using Neighborhood Path Filtering Clustering Algorithm

Contact Info

Product

Resources

About