Abstract:Content-based stereo video and image retrieval relies on traditional pattern feature extraction and cultural creative design applications. In a 3D system, the 3D model is typically projected into a 2D point cloud, after which feature extraction takes place. A depth map is an image created by orthogonal projection of three-dimensional data or viewpoint matching, parallax calculation, and resampling into regular data based on depth value. In this paper, a multilevel histogram-based shape segmentation method is proposed. The threshold can be determined by analyzing the relationship between the peaks if there are several obvious peaks on the image histogram. This paper focuses on the use of traditional pattern feature extraction and cultural creative design based on the shape segmentation of multilevel histograms. Each algorithm can achieve good classification results in the experiment of rotation invariant similarity classification. The average accuracy of the HOD algorithm is 84.1%, the average accuracy of the RSDF algorithm is 86.2%, the average accuracy of the GIF algorithm is 81.3%, and the average accuracy of this paper's algorithm is 89.9% when the rotation angle is 60–120°. The results show that this paper's algorithm has a higher classification accuracy compared to the other algorithms and that it maintains a high average classification accuracy across all rotation angles, indicating that it is rotation invariant and robust. The histogram-based threshold method has the advantage of not requiring knowledge of the image's prior information. The algorithm is easy to understand, straightforward, and quick. However, because the valley point in the histogram is not visible in images with low or high contrast between the target and the background area, it is difficult to select the threshold. The results of traditional pattern feature extraction can be improved using the retrieved pixels.

Generalised Vertical Projection Histograms Using Multi‐plane Homology

Multiple Pedestrians Tracking Based on Gaussian Probability Model

Random Walks Colour Histogram Modification for Human Tracking.

A Novel Method For Real-Time Object Detection And Multiple Persons Tracking

Multi-view Visual Surveillance and Phantom Removal for Effective Pedestrian Detection.

Multi-view Pedestrian Detection Using Statistical Colour Matching

Pedestrian detection based on modified dynamic background using gaussian mixture models and hog-svm detection

Robust Localisation of Pedestrians with Cast Shadows Using Homology in a Monocular View

Multi-view and Multi-Plane Data Fusion for Effective Pedestrian Detection in Intelligent Visual Surveillance

Partially Occluded Head Posture Estimation for 2D Images Using Pyramid HoG Features

Moving Shadow Detection Via Binocular Vision and Colour Clustering

Multiview Pedestrian Localisation Via A Prime Candidate Chart Based On Occupancy Likelihoods

Video-Based Facial Expression Recognition Using Histogram Sequence of Local Gabor Binary Patterns from Three Orthogonal Planes

Traditional Pattern Feature Extraction and Cultural Creative Design Application Based on Multilevel Histogram Shape Segmentation

Multiview Detection with Cardboard Human Modeling

Real-time Detection Via Homography Mapping of Foreground Polygons from Multiple Cameras.

A Hierarchical Algorithm for Human Posture Recognition Based on Spatial and Frequency Domain Features

Local Gabor Binary Pattern Histogram Sequence (LGBPHS): a Novel Non-Statistical Model for Face Representation and Recognition

Background Modeling from a Free-Moving Camera by Multi-Layer Homography Algorithm.

Local Histogram of Figure/Ground Segmentations for Dynamic Background Subtraction.

ANOMALY DETECTION AND LOCALIZATION: A NOVEL TWO-PHASE FRAMEWORK BASED ON TRAJECTORY-LEVEL CHARACTERISTICS