Computer Informatics Project: Implementing Search Techniques

This project report delves into the realm of computer informatics, focusing on the implementation and analysis of various search techniques for image retrieval. The project encompasses three primary tasks: developing a controlled vocabulary for image indexing, implementing a keyword-based ...Read More

Computer Informatics

Contents
Contents................................................................................................................................................2
1. Introduction...................................................................................................................................3
2. Task 1............................................................................................................................................3
2.1 Controlled vocabulary............................................................................................................3
2.2 Indexing Image......................................................................................................................3
2.3 Meta data information vocabulary.........................................................................................4
3. Task 2............................................................................................................................................4
3.1 Keyword Based Search..........................................................................................................4
4. Task 3............................................................................................................................................5
4.1 Image Based Search...............................................................................................................5
5. Spatial analysis..............................................................................................................................5
5.1 Spatial Characterization.........................................................................................................6
6. Strength and weakness of keyword based search...........................................................................6
7. Strength and weakness of image based search...............................................................................6
8. Conclusion.....................................................................................................................................7
References.............................................................................................................................................8

1. Introduction
Creating the image based on information that is communicating with the computer,
provides the related images to the user. The image can access the keyword components of the
user. The user searches the image based on the keyword. The Metadata can be used by the
user to provide information about the other data (Gaikwad et al., 2015). There are different
types of Metadata that are used and some of them are, the structural metadata, image retrieval
metadata and image preconization metadata, which will be investigated in this report.
This report aims to understand the performance and limitations of the salient
algorithms associated with different aspects of informatics. To understand the case studies,
practical application of the modern informatics techniques to various domains.
This report's objective is to study the computer informatics service platform. This
platform manages large-scale information and compiles a collection of images, used for the
query of Metadata associated with the images. Here, the python code will be used for the
implementation.
2. Task 1
2.1 Controlled vocabulary
This is the first task. Here, the task undertakes, developing a controlled vocabulary.
The image is used to implement the vocabulary. The abstract scene dataset contains scene
images. The images reflect on the children. First, it creates a description for every scene. The
vocabulary tools are used to find the vocabulary for the scene. The controlled vocabulary is
used for indexing schemes, and subject headings of the organization system. The image is
indexed by the content based retrieval system, depending on the controlled vocabulary. For
every word, vocabulary is created. It describes all the single words and is used to search the
keyword. In this task, it contains 58 image tags. The image tags convert the vocabulary
(Kulkarni, 2012).

2.2 Indexing Image
The image indexing depends on the text-based, description-based and content-based
image retrieval. For image indexing, the content-based image retrieval system software is
used. Image indexing means to display the description of the images. The image indexing
depends on the text-based, description-based and content-based image retrieval. For image
indexing, the content-based image retrieval system software is used. Image indexing means to
display the description of the images. Any image is searched using a particular keyword. The
keyword depends on the image description. The description of all the words is developed in
the first task. The description helps image retrieval. The image folder imports by using the
retrieval software and it displays the image with a description. The image represents the
colour, shape, and text. It has four types of approaches, where the first approach depends on
the attribute. The second approach depends on the object recognition system. The third
approach depends on image annotation. Here, the third approach is a feature of the low-level
image (MURALIDHARAN and POKHAREL, 2012). The above four approaches depend on
image indexing.
2.3 Meta data information vocabulary
For keyword search of Metadata vocabulary information, the English language is
used. It develops the definition of the keyword vocabulary system, which is used for image-
searching. It allows the inexperienced users to develop the custom metadata vocabularies to
suit their needs, to modify and merge the image to the keyword, and the keyword to image on
the existing vocabularies. Then, it associates the vocabularies with the specific metadata and
searches the vocabularies with the other users. The Meta information is used to have an
understandable machine structure, related data. In the vocabulary, the community of the
keyword and image, are stored in Metadata query format, which is often used for catalog
keyword. The image records contain the descriptive information. The bibliographic type

information can be described based on the selected index headings vocabulary (Patel et al.,
2013). The user can access information resources, for different types of methods and
functions. The Metadata information follows two types such as,
1) Image search metadata
2) Keyword search metadata
3. Task 2
3.1 Keyword Based Search
The keyword metadata is used for one of the most useful section of the vocabulary
format. It allows the user to search the words with the keyword using “cap”, and this
identifies the word on the query system, then it displays the word based on the “capacity” of
the vocabulary.
The keyword search engine is only used for searching the assigned keywords and
displays the full form of the text words. When compared to the full-text search engines, it is
much less and valuable (Wei, Zhang, and Lu, 2016). While using the full-text search engine,
it is used to improve the search rankings that are best in the whole world. The keyword search
looks for the words in the records. The user can search the word on the server that the
keyword searches and retrieves a large number of word results. The keyword is a word or
phrase, which is identified as a potential user while performing the searching operation on the
internet.
4. Task 3
4.1 Image Based Search
Designing the Metadata vocabulary management refers to maintaining the
keywords and images on the vocabularies which include, Metadata repository,
database, and the query. The Metadata information can follow the three types of
techniques such as,
1) Descriptive metadata
2) Structural Metadata
3) Administrative metadata

The descriptive Metadata can be used for finding the images. The user can search the
keyword vocabulary, by using the search engine on the query and then find the
specific words to be searched, retrieved, and managed. Finally, it displays the
keyword based images which include, word creator, image creator, creation date,
keyword, and basic identification information.
Structural Metadata can be used for the structure of image files that are the
interconnection for the inner parts of the database and allows the keyword to be
searched and presented. The administrative Metadata can be used for searching all the
information. Thus, it must be secured and maintained in the database.
5. Spatial analysis
The spatial analysis is followed by the three properties that include,
topological, geometric and geographical properties. The spatial analysis is used for
keywords, related to occupying the character on space, involving the perception of
relationships in the space tests related to the spatial ability’s spatial memory.
5.1 Spatial Characterization
The statistical characterization technique favours the spatial definition of the
objects as points. Because, there are very few statistical techniques which operate
directly online, area, or volume elements. The computer tools favour the spatial
definition of the objects as homogeneous and separate the elements, because of the
limited number of database elements and computational structures available, and the
ease with which this primitive structure can be created.
6. Strength and weakness of keyword based search
Strength
 By using this method a specific image is searched.
 This method contains two research communities they are, computer vision and
database management.
 This method is more effective and is very less expensive.
 The time to save for search.

Weakness
 It does not have accuracy.
 Continuous filtering is needed for accurate output.
7. Strength and weakness of image based search
Strength
 To process the reverse image search engine, the relevance of indexing images can
be classified.
 We can introduce a threshold for each similarity measure, to precisely select the
image that can be considered as similar to the query image.
 It is important to measure the time taken by the system, for indexing and
searching.
 Instantly, the system should be able to index millions of images and search across
them.
 The complexity of both the indexing and searching phases depends on the number
of images and it is not necessarily linear.
Weakness
 The multiple images cannot be searched at the same time. Because an error will
occur.
 It does not display the details of the images.
8. Conclusion
Implementation of controlled vocabulary, keyword-based search, and image-based
search is undertaken. The first task implements the vocabulary for the image tags. The
vocabulary and description are created by using the vocabulary tool and the vocabulary insert
is used for image indexing. The image indexing method displays the image description. The
second task implements the keyword-based search engine. The dataset imports using the
python code and it displays the image-related keyword. The input method is text-based and
the output method is image-based. The third task implements the image-based search engine.
It is a reverse method of the keyword-based search engine. The input is image-based and the
output is text based. The image imports the search engine box and it displays the description
for this image. Finally, the project is implemented successfully.

References
Gaikwad, P., Sanap, S., Jadhav, R., Kaware, S. and Mhasawade, P. (2015). ADVANCED
TECHNIQUE TO RETRIEVE FACE IMAGE BY USING ATTRIBUTE-ENHANCED
SPARSE CODEWORDS & INVERTED INDEXING. International Journal of Advance
Engineering and Research Development, 2(03).
Kulkarni, P. (2012). SD-miner System to Retrieve Probabilistic Neighborhood Points in
Spatial Data Mining. IOSR Journal of Computer Engineering, 4(6), pp.01-05.
MURALIDHARAN, B. and POKHAREL, R. (2012). Automatic Side Stand Retrieve
System. Paripex - Indian Journal Of Research, 3(2), pp.114-115.
Patel, V., Kwok, J., Sproat, C. and McGurk, M. (2013). To retrieve or not to retrieve the
coronectomy root – the clinical dilemma. Dental Update, 40(5), pp.370-376.
Wei, Y., Zhang, J. and Lu, Z. (2016). A Novel Successive Cancellation Method to Retrieve
Sea Wave Components from Spatio-Temporal Remote Sensing Image Sequences. Remote
Sensing, 8(7), p.607.