Module CO3091/7091 Computational Intelligence and Software Engineering Problem Solving Submission date: Completed report must be submitted on Blackboard no later than 19.00, Tuesday 25th November, 2021. 1. Aims This assignment will be assessed (20% of the final mark of the module). Students are invited to prepare a short report that describes an approach of detecting objects in images. For example, someone may wish to develop a technique to detect humans in 2D or 3D images. Others may expect to develop an approach to detect brain cancer areas in 2D or 3D MRI images. The proposed approach may be part of the learning, exercising or exploring that students have experienced inside and outside the class. This provides students with an opportunity to increase their knowledge and understanding, and to enhance their skills in computational intelligence and software engineering. This coursework also allows students to consolidate and practise the knowledge, understanding and analysis that they have gained inside and outside the class. To help students prepare for this submission, a tutorial will be given in November, 2021. The tutorial video will be available on Blackboard afterwards. 2. Learning outcomes ? Knowledge o A range of object detection approaches. ? Skills o To properly use the literature. o To develop effective information collection abilities. o To prepare and deliver documentations for the specified task. o To work towards a task oriented deadline. ? Understanding o Analysis of a real-world problem. o Collecting materials and references for real-world problem solving. o Development of machine learning and optimisation solutions. 3. Key indicators ? Reading and writing independently. ? Use of documentation tools, such as Microsoft Word 2010 or newer. ? Clear structure and convincing arguments in the documentation. ? Proper use of references from the literature. 4. Suggested procedure ? To retrieve necessary references in the literature. ? To read through the references. ? To plan the structure of the report. ? To organise the description/statement of each section. ? To revise and improve the report. ? To submit the report online (Blackboard). 5. Report requirements The structure of the report can be found in Appendix 1. Text font: Ariel, 11 or larger. Three compulsory components detailed below must be implemented. Compulsory Components 1) Introduction. Less than 100 words. 2) Proposed technology. Less than 300 words. 3) Conclusion. Less than 100 words. Necessary appendices can be added to the end of the report without any page limit, which help the reader to better understand the report. If a compulsory component is missing, 15% - 70% marks will be deducted from the overall marks, depending on the missing components. 6. Project submission Please submit the entire report in the PDF format to Blackboard on or before 19.00 on 25th November, 2021. Regarding late submission, plagiarism and mitigating circumstances, please check the study guide of the module on Blackboard. 7. References (these are examples) [1] Wang, Y., et al. RODNet: Object detection under severe conditions using vision-radio crossmodal supervisor, https://arxiv.org/abs/2003.01816. [2] Hinzmann, T., et al. Deep learning-based human detection for UAVs with optical and infrared cameras: System and experiments, https://arxiv.org/abs/2008.04197. [3] Hulzebosch, N., et al. Detecting CNN-generated facial images in real-world scenarios, https://arxiv.org/abs/2005.05632. [4] Li, Y., et al. Improving object detection with selective self-supervised self-training, https://arxiv.org/abs/2007.09162. [5] Zhang, Z., et al. A novel and efficient tumor detection framework for pancreatic cancer via CT images, https://arxiv.org/abs/2002.04493. [6] Huang, J., et al. Detection of human faces using decision trees, https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.51.4261&rep=rep1&type=pdf. 8. Marking scheme Your score depends on how well your report meets the following requirements: Percentage Section Classification 15% Introduction o 70% or above: Clearly explain the value and importance of object detection. Convincing explanation of why you propose your technology. o 60%-69%: Sufficiently explain the value and importance of object detection. Proper explanation of why you propose your technology. o 50%-59%: Acceptable explanation of the value and importance of object detection. Acceptable explanation of why you propose your technology. o 40%-49%: Less convincing explanation to the value and importance of object detection and why you propose your technology. o 39% or less: Poor explanation to the value and importance of object detection and why you propose your technology. 70% Proposed technology o 70% or above: Clearly introduce your approach by discussing each major component of the system. Well presented arguments throughout the section. o 60%-69%: Sufficiently introduce your approach by discussing each major component of the system. Reasonable arguments throughout the section. o 50%-59%: Acceptable explanation to your proposed object detection approach with some supportive statements. o 40%-49%: Less convincing explanation to your proposed object detection approach with limited supportive statements. o 39% or less: Poor explanation to your proposed object detection with limited supportive statements. 15% Conclusion o 70% or above: Convincingly summarise the proposed object detection technology with a convincing future plan. o 60%-69%: Sufficient summarisation the proposed technology with a sound future plan. o 50%-59%: Acceptable summarisation of the proposed object detection technology with a plan. o 40%-49%: Less convincing summarisation of the proposed object detection technology. o 39% or less: Poor summarisation of the proposed object detection technology. Appendix 1: Structure of the report recommendation but not limited to! Section 1 Introduction Why is object detection important? People have developed a number of technologies, why will they use your approach? Section 2 Proposed technology How does your proposed approach work in general? How does each component of your system work? Section 3 Conclusion Summarise the proposed object detection approach with remaining challenges to be solved to your understanding.

Attachments:

computational....pdf