Efficient <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si2.svg"><mml:mi>k</mml:mi></mml:math>-nearest neighbor search based on clustering and adaptive <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si2.svg"><mml:mi>k</mml:mi></mml:math> values

Gallego, Antonio‐Javier; Rico-Juan, Juan Ramón; Valero-Mas, Jose J.

doi:10.1016/j.patcog.2021.108356

Cited by 53 publications

(17 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The calculation of centroid coordinates is done by adding the coordinates of all points in the group and then dividing by the number of points. 8 The results are marked with green marker points on the images, as shown in Figure 7 and Figure 8.…”

Section: Centroid Fitting Of Marker Pointsmentioning

confidence: 99%

360 degree unique and robust matching method for the infrared optical features of handheld target

Duan,

Song,

et al. 2024

Speckle 2023: VIII International Conference on Speckle Metrology

View full text Add to dashboard Cite

Portable optical target measurement systems have widespread applications in various fields, including mechanical manufacturing, aerospace, industrial inspection, and clinical medicine. This study aims to address the marker point matching problem in binocular measurement systems. Through image preprocessing and mathematical algorithms, we have successfully achieved robust matching of marker points with uniqueness. We first converted color images to grayscale and applied the Otsu algorithm to adaptively select a global threshold, successfully extracting marker points. Using the nearest neighbor algorithm, we simplified 28 marker points to 7. Then, we employed the k-means clustering algorithm to achieve a second fitting, obtaining marker points located on the central line of the target. With these marker points, we computed key features for describing the target pose, including the direction vector of the target's central line and its angle with respect to the image's x-axis. Finally, we designed different algorithm modules to successfully achieve the robust matching of marker points across 360 degrees of free postures.

show abstract

Section: Centroid Fitting Of Marker Pointsmentioning

confidence: 99%

360 degree unique and robust matching method for the infrared optical features of handheld target

Duan,

Song,

et al. 2024

Speckle 2023: VIII International Conference on Speckle Metrology

View full text Add to dashboard Cite

show abstract

“…Related Work: State of the art can be roughly grouped as graph-based [16ś 19], hashing-based [20ś22], and partition-based [23,24] methods. The Faiss library [8] enables efficient partitioning of data in Voronoi cells [8], where the index of each cell is a centroid of that cell and product quantization is used [25] to compress data.…”

Section: State Of the Artmentioning

confidence: 99%

Stratified Graph Indexing for Efficient Search in Deep Descriptor Databases

Rahman,

Tešić

2023

Preprint

View full text Add to dashboard Cite

Searching for unseen objects in extensive visual archives is challenging, demanding efficient indexing methods that can support meaningful similarity retrievals. This research paper presents the Stratified Graph (SG) approach for indexing similar deep descriptors by sorting them into distance-sensitive layers. The indexing algorithm incrementally constructs a bi-directional m-nearest neighbor graph within each layer, with additional 1-nearest neighbor links from outer layers, providing a distant scaling property in the graph structure. The search process starts from the innermost layer, and the same layer neighbors enhance Average Recall (AR), while the distant scaling property enhances search speed, maintaining logarithmic complexity scaling. We compare and contrast SG with six state-of-the-art retrieval methods in four deep-descriptor and two classical-descriptor databases, and we show that the SG indexing and search has smaller memory usage (up to four times) and the Mean Average Precision and AR improve up to 8% over state-of-art for all six datasets at five retrieval depths.

show abstract

“…The algorithm presented in [17] cannot be considered a PG algorithm. However, it is able to perform efficient Nearest Neighbor searches in the context of k-NN classification.…”

Section: Related Workmentioning

confidence: 99%

Fast Training Set Size Reduction Using Simple Space Partitioning Algorithms

et al. 2022

View full text Add to dashboard Cite

The Reduction by Space Partitioning (RSP3) algorithm is a well-known data reduction technique. It summarizes the training data and generates representative prototypes. Its goal is to reduce the computational cost of an instance-based classifier without penalty in accuracy. The algorithm keeps on dividing the initial training data into subsets until all of them become homogeneous, i.e., they contain instances of the same class. To divide a non-homogeneous subset, the algorithm computes its two furthest instances and assigns all instances to their closest furthest instance. This is a very expensive computational task, since all distances among the instances of a non-homogeneous subset must be calculated. Moreover, noise in the training data leads to a large number of small homogeneous subsets, many of which have only one instance. These instances are probably noise, but the algorithm mistakenly generates prototypes for these subsets. This paper proposes simple and fast variations of RSP3 that avoid the computationally costly partitioning tasks and remove the noisy training instances. The experimental study conducted on sixteen datasets and the corresponding statistical tests show that the proposed variations of the algorithm are much faster and achieve higher reduction rates than the conventional RSP3 without negatively affecting the accuracy.

show abstract

Efficient k-nearest neighbor search based on clustering and adaptive k values

Cited by 53 publications

References 66 publications

360 degree unique and robust matching method for the infrared optical features of handheld target

360 degree unique and robust matching method for the infrared optical features of handheld target

Stratified Graph Indexing for Efficient Search in Deep Descriptor Databases

Fast Training Set Size Reduction Using Simple Space Partitioning Algorithms

Contact Info

Product

Resources

About