7COM1084 Advance Research Topic in Computer Science

Research Area

A research study is proposed for analyzing the technique of Image Captioning using Deep Learning Methods. Image captioning is the procedure of generating captions or textual description of a particular image linked with the contents of the Image. This process is mainly based on the technique of machine learning and deep learning. Through the use of Deep Learning techniques, automatic image captioning is possible (Hossain et al. 2019). Through this research study, a detailed analysis on how deep learning techniques and algorithms can be used for effective image captioning, will be presented.

Although it looks simple, Image captioning is a seemingly complex task. Image captioning is based on generation of output that will describe a particular system and their properties along with the interactions that is being carried out between the objects. Replication of this behavior in an AI enabled system is a huge task which can be simplified using the tools and techniques offered by machine learning and deep learning (Yao  et al. 2017). The aim of the proposed research study is to outline the process in which deep learning can be implemented for successful image captioning.

Relation to Computer Science In General

The topic chosen for the research is directly linked with computer science as Image captioning is possible for the systems that operates on the technology of AI. Artificial Intelligence is an emerging technology that has the capability of replicating human intelligence and perform tasks that required human intervention. AI enabled machines are programmed to perform works like human (Cui  et al. 2018). It is supported by tools such as deep learning and machine learning. Therefore, it can be said that the topic that is opted for this research study is directly linked to computer science in general as it is directly linked with the concept of AI which is an emerging topic and an application of computer science.

Image captioning has received significant importance over the last few years and hence several new methods are being proposed in order to achieve the expected and satisfactory results in this field. Deep Learning is a technique that is capable of improving the overall process of image captioning (You  et al. 2016). Through this research study a thorough analysis of how deep learning can help in image captioning will be addressed.  

Research Questions and Methods

The proposed research study aims in gaining answers to the following research questions-

1. What is the concept of Image captioning and How it works?
2. What is the role of Artificial Intelligence in development of an image captioning system?
3. How can deep leaning help in effective Image Captioning?
4. What are some of the strategies that can be implemented to improve the process of image captioning through the use of Deep Learning techniques?

The proposed research study aims in finding solutions to the above indicated research questions. The methodology that will be applied for the above indicated research questions are discussed in the following section.

The use of a quantitative research method is proposed for the particular research study that involves active experimentation to develop an image captioning tool using the technology of Depp learning.

For this research study the use of deductive research approach is proposed, as the proposed research study is directly linked with a scientific investigation on the use of deep learning technology for image captioning. Along with that the use of a positivism research philosophy is proposed for this particular study. Positivism research philosophy is appropriate for this research study as in this proposed work, the researcher will work independently to find solution to the identified research problem and the research questions. The use of positivism research approach will contribute in easier identification and the analysis of the overall experimentation and primary data collection processes to be carried out in this proposed research.

In this proposed research study, collection of data from both primary and secondary sources is suggested. The secondary data will be collected from the researches that has been already conducted in the field. The primary data will be carried out through an experimentation process to test how deep learning can help in designing an image captioning system and whether the use of Deep Learning Technology can contribute in automatic image captioning.

A basic methodology that can be considered for carrying out the experimentation process related to use of Deep Learning in Image Captioning are presented as follow-

First of all, a choice of a good dataset is recommended. A realistic dataset is necessary to test if Deep Learning can be effectively used for image captioning. The preparation of the photo data on which the entire experimentation process will be carried out is a necessity. In this experiment, the use of tools such as Keras is proposed which will contribute in easier experimentation (Brownlee 2019). Then the image captioning model is needed to be coded using deep learning algorithm. Once the coding is complete, the model is required to be configured for the testing purpose.

Along with the detailed experimentation process, primary data can be collected from the coders that have a first-hand experience of the use of deep learning in overall image captioning process. The tool that is proposed for data collection is the tool of interview which will contribute in having a clear understanding of how Deep Learning can be put to use in designing an effective image captioning tool. The data obtained from the interview will be a great help in the experimentation process as well since it will provide the necessary data on how deep learning algorithms can be used to design an image captioning system. The collection of accurate data from primary and secondary sources will increase the chances of success in the experimentation process. The data from the interview will be collected from 10 respondents chosen on basis of random sampling method. The data collected from interview and secondary sources will undergo qualitative data analysis.

Artificial Intelligence

Artificial intelligence is referred to as the replication of human intelligence in machines. Through this technology, the machines are programmed to think like humans and are also capable of mimicking human actions (Jackson 2019). It is an emerging topic which help in developing smart machines that have the capability of learning from experience and is capable of problem solving (Flasiński 2016).

Cyber Security

Cyber security can be described as the practice of defending computers, network, servers and storage devices from different types of malicious attacks (Gupta 2018). It encompasses the technology and the procedures that are intended to defend networks and devices from different types of attacks, damages and even from unintended access.

References

Brownlee, J., 2019. How to Develop a Deep Learning Photo Caption Generator from Scratch. [online] Machine Learning Mastery. Available at: <https://machinelearningmastery.com/develop-a-deep-learning-caption-generation-model-in-python/> [Accessed 3 May 2021].

Cui, Y., Yang, G., Veit, A., Huang, X. and Belongie, S., 2018. Learning to evaluate image captioning. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5804-5812).

FlasiÅ„ski, M., 2016. Introduction to artificial intelligence. Springer.

Gupta, B.B. ed., 2018. Computer and cyber security: principles, algorithm, applications, and perspectives. CRC Press.

Hossain, M.Z., Sohel, F., Shiratuddin, M.F. and Laga, H., 2019. A comprehensive survey of deep learning for image captioning. ACM Computing Surveys (CsUR), 51(6), pp.1-36.

Jackson, P.C., 2019. Introduction to artificial intelligence. Courier Dover Publications.

Yao, T., Pan, Y., Li, Y., Qiu, Z. and Mei, T., 2017. Boosting image captioning with attributes. In Proceedings of the IEEE International Conference on Computer Vision (pp. 4894-4902).

You, Q., Jin, H., Wang, Z., Fang, C. and Luo, J., 2016. Image captioning with semantic attention. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4651-4659).

Our team consists of PhD stalwarts who leave no stones unturned to deliver you perfect assignment writing services. The ghostwriters follow your university guidelines to ensure you get top-notch quality work irrespective of the topic. Our team has gone through rigorous training to meet your expectations with the utmost precision. We are familiar with all types of academic documents such as essays, academic posters, book reports, dissertations, homework, coursework and more. Whether you opt for our online exam help or assignment help, we provide individualised attention to your concern in each case.