Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi ka Zeyuan Zhang and Xuechun Tao

Moralo oa Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi ka Zeyuan Zhang and Xuechun Tao ke mohlodi oa khau e hlomphehang haholo ya 2025 Moputso oa Boronse oa A' Design Award Mokhahleong oa Tlhodisano oa Moralo wa Sebopeho, Tshebedisano le Boiphihlelo ba Mosebedisi.

BoronseSebopeho· 2025

Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi

ka Zeyuan Zhang and Xuechun Tao

Moreki: Tick Design Labs (Shanghai)

Mohlodi oa Moputso oa Boronse oa A' Design Award Mokhahleong oa Sebopeho

28,101

Mohlodi oa Khau ea Netefaditsweng

Netefatsa Boemo ba Khau bakeng sa #165869

Etela Sebaka Sheba ho A' Design Award Tokollo ya Boraditaba Baeche ya Moputso

Mebala

Moralo

Mabapi le Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi

Thermal Sense ke sesebediswa sa ho beha leihlo ho mokgweetsi bakeng sa dikoloi. Thermal Sense e nkela sebaka sa khamera e tlwaelehileng ya ka hare ho koloi ka sensara e tlase ya resolution ya mocheso, e hlophisang feela mekgwa ya mocheso, kahoo ha ho ditshwantsho tse ka tsejwang tse etswang. Sekoloto sa sensara se lekana dimilimithara tse 40 ka tse 30 ka tse 20 mme se hlomilwe ho seiponeng sa ka morao bakeng sa tshireletso e batsi ya ka hare ho koloi. Marangrang a methapo a hlahloba maemo a ho kwala mahlo le motsamao wa hlooho ho fumana ho robala, ho shwa dikelellong le boteng ba ngwana. Metsotswana ya tshireletso e fanwa nakong e ka tlase ho 100 ya milliseconds, mme lekunutu la data le sireletswa maemong a data.

Kamohelo

Mabapi le Moputso oa Boronse oa A' Design Award

Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi ke mohapi oa Moputso oa Boronse oa A' Design Award, kananelo e hlomphehang haholo e fuoang meralo e ikhethang e tsejoang ka boqapi, bohlale ba ho qapa le bohlale. Bahapi ba Boronse ke lintho tse nang le bokhoni ka ho felletseng le boqapi bo hlollang tse tiisang boiphihlelo bo bongata, mohopolo le bokhoni ba ho iphumana ba baqapi ba tsona. Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi e lebelletsoe ho ba ntho e hlomphuoang haholo, e ratoang le e ratehang, e nang le boprofeshenale bo bonahalang le mehopolo e mecha. Moralo ona o hapileng Moputso oa Boronse oa A' Design Award o kenyelletsa mekhoa e metle ka ho fetesisa bonong, mahlaleng, moralong le theknolojing mme o hlahella ka ho tsepamisa maikutlo lintlheng tse nyenyane. Thermal Sense Sesebediswa sa ho Beha Leihlo ho Mokgweetsi e kutloisisa litlhoko hantle, e ama litakatso, e hapa maikutlo mme e tsosa takatso ea ho tseba.

Ka Mora Moralo

Sete ya Boraditaba

Litlhaku, ditho tsa sehlopha le lintlha tsa botekgeniki tse fanoeng ke moetsi wa moralo.

Boleng bo Ikhethang: With the increasing requirement for in-car cameras to monitor drivers and support assisted driving systems, Thermal Sense offers a groundbreaking alternative. It replaces traditional cameras with low-resolution thermal sensors, capturing only heat signatures to eliminate privacy risks. This system ensures data-level privacy while detecting drowsiness, distractions, and child presence using advanced machine learning. By prioritizing safety and privacy, Thermal Sense sets a new benchmark.
Boholo le Phuthelo: The thermal sensor module measures 40mm x 30mm x 20mm and is mounted discreetly on the rearview mirror for optimal cabin coverage. It features a low-resolution array of 32x32 pixels with a thermal sensitivity of ± 0.1°C. The system operates in real-time with a latency of under 100ms, powered by a custom hardware platform integrating convolutional neural network architectures. The package is lightweight, energy-efficient, and optimized for seamless automotive integration.
Lithegi: thermal imaging, privacy monitoring, driver analysis, cabin safety, thermal sensors, non-invasive, AI
Theknoloji ya Tlhahiso: Thermal Sense integrates a thermal sensor at the rearview mirror for wide-angle cabin coverage. Sourced with customized original equipment manufacturer specifications, the sensor detects minute thermal variations, ensuring high sensitivity for monitoring physiological parameters. Advanced convolutional neural network process thermal data, analyzing key indicators like eye closure rates and head movements. A robust real-time pipeline enhances image quality, extracts features, and provides immediate safety alerts.
Phephetso ya Moralo: Limited availability of high-quality, low-cost thermal sensors and real-time processing technology were overcome through iterative testing and collaboration with original equipment manufactures and artificial intelligence experts.
Dipatlisiso: Using an in-between method, we cross-compared the thermal sensor approach with traditional cameras for measuring driver behavior. Controlled experiments with 50 participants assessed drowsiness and distractions. Results showed no significant performance difference between the two methods. This demonstrates that the thermal approach is as effective as cameras while eliminating privacy concerns, making it a superior, privacy-conscious solution.
Kgothaletso: Inspired by advancements in thermal imaging and the need for privacy-focused solutions, Thermal Sense addresses mandatory driver monitoring requirements without compromising identity. Research shows the efficacy of thermal sensors in detecting behavior and ensuring safety. By eliminating human-identifiable color images, it prioritizes privacy while monitoring driver attention, passenger activity, and child presence.
Nako ya Projeke: The project commenced in January 2023 in Shanghai, China, and was completed in October 2024. Research, development, and testing phases were conducted in collaboration with leading automotive technology innovation labs.
Tsamaiso ya Mosebetsi: Thermal Sense operates seamlessly in the background, using a thermal sensor to monitor the cabin without capturing identifiable images. Installed on the rearview mirror, it detects heat patterns to identify driver drowsiness, distractions, or child presence. The system provides real-time alerts, ensuring safety without compromising privacy. Its unobtrusive design and precise detection deliver reliable results, creating a safer and more privacy-conscious driving experience.
Diphatente: Copyrights belong to Zeyuan Zhang, and Tick Design Labs (Shanghai)
Litho tsa Sehlopha: Lead Designer: Zeyuan Zhang
Lead Designer: Xuechun Tao
Lead Designer: Yangyi Tai
Litlotla tsa Setshwantsho: All Image: Zeyuan Zhang