Experimental Log

The following table is a record of the experiments we carried out in order to test some of the functionality of the system.

API

Experiment # Title Details Results Success Done By Date
1 Data Integrity Certain data of models should follow specific database rulings, such as ‘uniqueness’ Hyojong Kim
1.1 Uniqueness of a coordinate and its location Inserted a pair of coordinates that already exist in the database to create an area code. Failed Yes Hyojong Kim 2 Nov - 9 Nov
1.2 Uniqueness of a user Created a user with an existing credentials Failed Yes Hyojong Kim 2 Nov - 9 Nov
1.3 Model Behaviours Created models and ran class functions Success Yes Hyojong Kim 11 Nov - 18 Nov
2 Server Performance Server should be able to handle multiple concurrent requests and in cases with millions of items Hyojong Kim
2.1 Multiple Queries Ran 100 concurrent queries using threading Failed No Hyojong Kim 11 Nov- 18 Nov
2.2 Multiple Items Had 4 mil items and expected the query time < 0.5 Success Yes Hyojong Kim 11 Nov - 18 Nov
3 API Behaviours API built should behave in an expected manner Hyojong Kim
3.1 Correct Parameters Made a get request to fetch an area code Success Yes Hyojong Kim 23 Nov - 30 Nov
3.2 Wrong Parameter Made a get request to fetch a user and an area code with wrong parameter Failed Yes Hyojong Kim 23 Nov - 30 Nov

PoC

Experiment # Title Details Results Success Done By Date
1 Camera and Face detection Open and use the front and back camera of multiple platforms to detect faces using Haar Classifier Vu Luong, Arjun Khurana 10 Nov – 16 Nov
1.1 macOS Sierra (Macbook Pro) Load the camera and face detection algorithm written in C++ in Xcode Front camera loaded successfully and detected the user’s face as well as faces from advertisements Yes Vu Luong 10 Nov
1.2 iOS 10.1.1 (iPhone 6 and 7+) Load the camera and face detection algorithm written in Objective C in Xcode Same as 1.1 Yes Vu Luong 12 Nov
1.3 Android 6.0 (Google Nexus 5) Load the application written in Java and JavaCV in Eclipse on to the device Failed to load the application onto the device after multiple attempts, possible cause due to faulty installation. Hence, Android is no longer the platform for development. No Vu Luong, Arjun Khurana 13 Nov – 16 Nov
2 Recognising shapes and texts Using OpenCV and Tesseract OCR Vu Luong, Hyojong Kim 20 Nov - 21 Nov
2.1 Shape detection The app should be able to recognize rectangular shapes using Harris corner detection Successfully recognise shapes of different sizes Yes Vu Luong 20 Nov
2.2 Text detection Using Tesseract OCR to recognise text on different surfaces Successfully recognise dark color text printed on paper and on well lit images. Takes a long time to recognise text with unique fonts, colours and sizes Yes Vu Luong, Hyojong Kim 21 Nov
3 Make request to server Making get/post requests that allow the app to have the database access Vu Luong, Hyojong Kim 21 Nov - 27 Nov
3.1 User profile Using User object received from customer to generate temporary(persistent) user profile Successfully requested user profile. Yes Vu Luong, Hyojong Kim 21 Nov - 24 Nov
3.2 Advertisement data Supplying coordinate from iPhone GPS system and receiving the intended area code and its advertisement GPS coordinate and area code retrieval success, advertisement sometimes unable to load No Vu Luong, Hyojong Kim 24 Nov - 27 Nov
4 Saving data Saving user profile data onto the local device and server Vu Luong, Hyojong Kim 24 Nov - 02 Dec
4.1 Device storage Use local databases such as SQLite to create a persistent database Successfully saved Yes Vu Luong 24 Nov - 02 Dec
4.2 Server Making post request based on user activities and save a history of user activities User activities recorded on server Yes Hyojong Kim 27 Nov - 02 Dec