Antropometri boleh ditakrifkan sebagai sains pengukuran dimensi saiz dan perkadaran tubuh manusia, sama ada hidup atau mati. Pengukuran antropometri digunakan dalam banyak bidang seperti perubatan, forensik dan pakaian. Teknik antropometri moden adalah mahal dan mengambil banyak ruang manakala teknik antropometri tradisional adalah lambat dan terdedah kepada kesilapan manusia. Kaedah anthropometri tradisional juga menimbulkan risiko gangguan seksual semasa pengukuran dijalankan manakala kaedah antropometri moden menggunakan sinar-X boleh menyebabkan kemudaratan kepada kesihatan dan alam sekitar manusia akibat radiasi. Oleh itu, timbulnya keperluan untuk membangunkan sistem anggaran berasaskan penglihatan yang selamat, murah, cepat, tepat dan mudah alih. Kajian dalam projek ini mencadangkan cara baharu menggunakan Raspberry 3B + dengan kamera Logitech C310 untuk melakukan anggaran lingkaran dada, pinggang, pinggul dan ketinggian badan melalui imej 2D. Sistem ini berfungsi dengan melaksanakan Kaedah Ketumpatan Pixel, kaedah ambang HSV, rangkaian neural dalam OpenPose, kaedah kotak terikat, kaedah pengiraan piksel dan kaedah anggaran lingkaran elips Ramanujan untuk mencapai objektif kajian ini. Perbezaan peratusan antara keputusan anggaran dan keputusan ukuran dada, pinggang, keliling pinggang dan ketinggian badan adalah hanya 2.11%, 4.66%, 4.31% dan 1.74%. Sebagai kesimpulan, sistem anggaran antropometri berasaskan penglihatan berjaya dibangunkan dengan ketepatan yang tinggi menggunakan semua kaedah yang dicadangkan.
_______________________________________________________________________________________________________
Anthropometry can be defined as the science of dimensional measurement of the size and proportions of the human body, whether living or dead. Anthropometric measurement is used in a lot of fields such as medical, forensics and clothing. The modern anthropometry techniques are expensive and takes up a lot of space while traditional anthropometry techniques are slow and susceptible to human error. Traditional anthropometry method also poses a risk of sexual harassment during measurement while modern anthropometry method using X-ray might cause harm to human health and environment due to radiation. Hence, there is a need for the development of a vision-based estimation system that is safe, cheap, fast, accurate and portable. The study in this project proposes a noble way of using Raspberry 3B+ with Logitech C310 camera to perform estimation of chest, waist, hip circumference and body height through 2D images. The system works by implementing Pixel Density Method, HSV thresholding method, OpenPose deep neural network, bounding box method, pixel counting method and Ramanujan’s ellipse circumference approximation method to achieve the objectives of this study. The percentage error between the estimated result and the measured result of chest, waist, hip circumference and body height are only 2.11%, 4.66%, 4.31% and 1.74%. In conclusion, a vision based anthropometric estimation system is successfully developed with high accuracy using all the proposed methods.
Development of a vision based anthropometric estimation system / Gan Xu Yang