Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Cognitive Internet of things : Frameworks, Tools and Applications
Lu H., Springer Publishing Company, Incorporated, 2019. 522 pp. Type: Book (978-3-030049-45-4)
Date Reviewed: Jan 7 2022

The introduction to this book cites its aim:

to collect a series of state-of-the-art contributions to the fields of computational neuroscience, computational cognition and perception, computer vision, natural language processing, human action analysis and related topics.

The result takes the form of 50 short papers--about ten pages each--on the topics. The way in which the papers are ordered is not made explicit, so readers will need to scan the titles to determine which are of interest. Sometimes related papers are next to one another, but not always. Given the number of papers in the book, it does not make sense to go into detail on each paper or even list all the papers. So this review will give an overview of the topics in somewhat greater specificity than that of the book’s introduction. An effort has been made to place related topics close together. The coverage is not exhaustive, nor are the topics in any particular order.

Object tracking is a topic that is the subject of several papers. Mostly the papers deal with the tracking of human beings, although one paper considers the case of determining which bees in a crowd are moving. Papers on face detection and enhancement of images of faces (called in one paper hallucination) fall into this general category. Underwater imaging and underwater image detection are the subjects of two papers. One could place a paper that deals with image enhancement of images made under foggy conditions in this group.

Moving to the topics of image enhancement and photography provides themes that appear in several papers. In addition to pure image enhancement, we find a paper that addresses the issue of inserting images into larger images in such a way that the shadows are consistent with the ambient lighting. Correction of distortion in zoom lenses also belongs in this grouping. One of these photography-related papers considers the automatic generation of film trailers.

Semantics is a consideration in several papers, and not just in the case of natural language processing. Two papers specifically address semantic issues for Chinese language systems. Another paper seeks to use semantics to improve hashing, so that semantically related items hash close to each other.

Aside from the paper on hashing alluded to above, there are (as one might expect) several papers on data retrieval. Multimodal retrieval where the request uses one mode (for example, an image) and the desired data is in another mode (text, for example) is the subject of one paper.

In terms of real-world applications, we find papers that deal with traceability in the pig supply chain and a study of bicycle availability in urban bicycle sharing. Two papers address the global positioning system (GPS) and the elimination of ambiguity in position determination. One paper amongst others related to medical topics considers the problem of detecting counterfeit medications. Finally in this grouping there is a paper that uses the ability to detect actual aircraft take-off times as support for “intelligent airports.”

There are several papers whose main content is the use of specific algorithms. Thus particle swarm algorithms are used in several papers, and an interesting search algorithm called the grey wolf algorithm is the topic of one of the papers in this category. While on the topic of algorithms, it should be noted that many of the papers use neural networks as the framework for their systems, in particular convolution neural networks. In most cases, the papers describe some special design element of the network for a particular problem.

As the preceding paragraphs indicate, the book covers a wide range of topics. The demands of the final form of the book mean that the papers are concise, even terse on occasion. However, in general, the structure of each paper allows the reader to determine, through the abstract, introduction, and future work sections, its applicability. Each paper has a comprehensive bibliography enabling the interested reader to pursue the topic further.

The book serves as an overview of a body of research into a wide range of topics. Despite what the title suggests, not every paper concerns the Internet of Things (IoT), at least not directly. “Frameworks” and “tools” appear--primarily neutral networks, denoising, and object enhancement--but not usually as explicit topics of any paper. A more comprehensive introduction, or an organization of the papers into sections, would make the book more easily accessible. In spite of this, graduate students will find the book useful as a snapshot of the research being carried out primarily by Chinese and Japanese researchers. This geographical leaning is not surprising given that the support for the project came from Japanese and Chinese organizations.

Reviewer:  J. P. E. Hodgson Review #: CR147398
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Internet (C.2.1 ... )
 
 
Artificial Intelligence (I.2 )
 
Would you recommend this review?
yes
no

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy