Multimedia Data Mining: A Systematic Introduction to by Zhongfei Zhang

By Zhongfei Zhang

Collecting the most recent advancements within the box, Multimedia facts Mining: A Systematic advent to thoughts and Theory defines multimedia info mining, its concept, and its purposes. of the main lively researchers in multimedia info mining discover how this younger sector has quickly built in fresh years.

The publication first discusses the theoretical foundations of multimedia info mining, proposing conventional characteristic illustration, wisdom illustration, statistical studying, and tender computing ideas. It then offers program examples that exhibit the good capability of multimedia information mining applied sciences. during this half, the authors express how you can improve a semantic repository education strategy and an idea discovery procedure in an imagery database. They show how wisdom discovery is helping in attaining the target of images annotation. The authors additionally describe a good way to large-scale video seek, in addition to an software of audio info category and categorization.

This novel, self-contained publication examines how the merging of multimedia and information mining study can advertise the certainty and develop the advance of data discovery in multimedia data.

Show description

Read Online or Download Multimedia Data Mining: A Systematic Introduction to Concepts and Theory (Chapman & Hall/CRC Data Mining and Knowledge Discovery Series) PDF

Similar imaging systems books

Morphological Image Analysis: Principles and Applications

From studies of the 1st variation: "This is a scholarly travel de strength throughout the international of morphological photograph research […]. i like to recommend this ebook unreservedly because the most sensible one i've got encountered in this specific subject […]" BMVA information

Handbook of Optical and Laser Scanning, Second Edition (Optical Science and Engineering)

From its preliminary ebook titled Laser Beam Scanning in 1985 to instruction manual of Optical and Laser Scanning, now in its moment variation, this reference has stored pros and scholars on the vanguard of optical scanning expertise. rigorously and meticulously up-to-date in each one generation, the booklet remains to be the main entire scanning source out there.

2D and 3D Image Analysis by Moments

Offers fresh major and fast improvement within the box of second and 3D photograph research second and 3D photograph research through Moments, is a different compendium of moment-based photograph research consisting of conventional equipment and likewise displays the newest improvement of the sector. The publication offers a survey of second and 3D second invariants with appreciate to similarity and affine spatial variations and to photograph blurring and smoothing via numerous filters.

Additional info for Multimedia Data Mining: A Systematic Introduction to Concepts and Theory (Chapman & Hall/CRC Data Mining and Knowledge Discovery Series)

Example text

Means there exists at least one value of the variable to which this quantifier applies. Given this FOL, the natural language sentence All the blue regions are either sky or water may be represented as the following FOL statement: ∀x(blue(x) → sky(x) ∨ water(x)) where blue(), sky(), and water() are the predicates, and → means “imply”. On the other hand, the natural language sentence Some blue regions are sky may be represented as the following FOL statement: ∃x(blue(x) → sky(x)) The advantage of using FOL for knowledge representation in a multimedia data mining system is that FOL makes deductive reasoning very easy and powerful; the reasoning process is also very efficient due to the symbolic computation using the FOL statements.

Consequently, given multiple modalities of multimedia data, we may use a feature vector to describe the data in each modality. , a concatenation of all the feature vectors for different modalities) © 2009 by Taylor & Francis Group, LLC Feature and Knowledge Representation for Multimedia Data 49 if the mining is to be performed in the whole data collection aggregatively, or we may leave the individual feature vectors for the individual modalities of the data if the mining is to be performed for different modalities of the data separately.

Given a regular histogram vector, data points in each component (called a bucket) of the histogram vector are further partitioned into two groups, one called coherent data points and the other called incoherent data points. A group of data points is defined as coherent if they are connected to form a connected component in the original domain of the multimedia data; otherwise, the data points are defined as incoherent. The specific implementation of the coherence definition is to set up a threshold c in advance such that a group of data points are coherent if their total count in the number of the data points that are connected exceeds c.

Download PDF sample

Rated 4.82 of 5 – based on 45 votes