全國中小學科展

2022年

提升戶外物件辨識模型表現之研究

近年來由於電腦視覺的蓬勃發展,物件辨識模型被廣泛運用在生活中,例如自動駕駛、醫療影像、農作物檢測等等。對於要在戶外運作的模型,由於檢測物體的背景會隨著時間、地點、季節、光照強弱等因素不斷改變,通常需要大量且多元的資料才能避免模型過擬和,然而取得多元的資料需要花費大量的人力與時間在收集以及為這些新資料標籤。 本研究利用影像風格轉換模型作為資料增強的方法,將於晴天拍攝的街景圖轉成夜晚及雨天,使原本只有晴天的資料集有更多元的資料。結果證實使用風格轉換模型生成的影像訓練的物件辨識模型的準確率在某些情況下有顯著的提升。此方法的優點在於能夠快速產生多樣風格的資料,由於是對影像的風格做轉換,影像的內容沒有改變,因此能夠沿用原有的標籤,同時節省了蒐集及為新影像標籤的人力及時間。

在壓力下更要堅強,探討ACE2/Ang(1-7)軸與韌力的關係

背景:韌力 (resilience)指個體遭受壓力時快速調適及復原的能力,是許多疾病的保護因子但生理機轉仍不明。Angiotensin-converting enzyme 2/Angiotensin(1-7) [ACE2/Ang(1-7)]軸具有保護心血管、抗焦慮及保護神經的潛力。目前仍無ACE2/Ang(1-7)軸與韌力相關的研究。另外ACE2是SARS-CoV-2病毒進入人體的受體,相關研究在疫情下更顯重要。 目的:本研究中我們以習得無助行為模式篩選具韌力行為的大鼠(resilient, R),探討韌力鼠與非韌力鼠(non-R) ACE2/Ang(1-7)軸表現的差異。 假說:韌力鼠的ACE2/Ang(1-7)軸活性較非韌力鼠高。 結果:韌力鼠在壓力下其corticosterone (CORT)濃度與控制組無差異。韌力鼠的Ang(1-7)濃度較非韌力鼠及控制組高,暗示較高的Ang(1-7)濃度與韌性表現相關。以Ang(1-7)/Angiotensin II (Ang II)比率作為反映ACE2 轉化Ang(1-7) 功能的指標,發現韌力鼠的ACE2濃度及功能均與未主動逃跑而受電擊的時間相關,暗示其ACE2活性能因應壓力調控上升。韌力鼠的ACE2濃度雖較非韌力鼠低,但其ACE2功能則較其他兩組高。並且,Ang (1-7)濃度及ACE2功能僅在韌力鼠與ACE2濃度呈現正相關,但在非韌力鼠則無相關。暗示在壓力下韌力鼠的ACE2仍維持有效率的功能但非韌力鼠的ACE2功能則受影響。另一方面,Ang (1-7)濃度僅在非韌力鼠與CORT濃度負相關,暗示非韌力鼠個體的ACE2/Ang (1-7)]軸活性與海馬迴-下視丘-腦下垂體(HPA)軸的活性可能相關。 結論:韌力個體的ACE2/Ang (1-7)軸活性較非韌力鼠及控制組佳且可能因應壓力而調節增加。相反的,非韌力鼠個體的ACE2/Ang (1-7)軸功能較差且可能與HPA軸的活性相關。我們的研究顯示ACE2/Ang (1-7)軸扮演調節韌力的重要角色。ACE2/Ang (1-7)軸活性是否可以做為個體韌力的指標或具有增進韌力的臨床價值需進一步研究。

腦波辨識特徵提取於即時身分認證的研究

本研究的腦波辨識基於特徵提取,可應用於身分認證,具有不能被仿冒的優點。我們用低成本高便利性的腦波儀,自行撰寫程式讀取原始腦波,建立一致性的實驗程序。首先用腦波專心度的高低來控制智能車,再用腦波來測謊,有隱藏說謊行為時會觸發高電位腦波,在兩項前期研究後發現可用腦波特徵進行身分認證。三位受測者於不同日期提取10份腦波,每份腦波紀錄5120筆數據。接著我們反覆嘗試組合數十種統計函數進行特徵提取,找到兩項最佳特徵,達成將大量凌亂腦波資料降低維度又具有辨識力。我們腦波辨識分類方法使用近鄰演算法,測試程序用盲測交叉驗證法,辨識正確率百分百。最後我們用Arduino板來展示腦波辨識應用於腦波身分認證,資料庫中只要儲存每位受測者的腦波特徵值,就能在數秒內正確辨識說出受測者身分,顯示每個人腦波是不同的,而且能用特徵將其分辨出來。

A.N.T.s: Algorithm for Navigating Traffic System in Automated Warehouses

According to CNN Indonesia 2020, the demand for e-Commerce in Indonesia has nearly doubled during this pandemic. This surge in demand calls for a time-efficient method for warehouse order-picking. One approach to achieve that goal is by incorporating automation in their warehouse systems. Globally, the market of warehouse robotics is expected to reach 12.6 billion USD by 2027 (Data Bridge Market Research, 2020). In this research, the warehouse system studied would utilize AMR (Autonomous Mobile Robots) to lift and deliver movable shelf units to the packing station where workers are at. This research designed a heuristic algorithm called A.N.T.s (Algorithm for Navigating Traffic System) to conduct task assigning and pathfinding for AMR in the automated warehouse. The warehouse layout was drawn as a two-dimensional map in grids. When an order is placed, A.N.T.s would assign the task to a robot that would require the least amount of time to reach the target shelf. A.N.T.s then conducted pathfinding heuristically using Manhattan Distance. A.N.T.s would help the robot to navigate its way to the target shelf unit, lift the shelf and bring it to the designated packing station. A.N.T.s algorithm was tested in various warehouse layouts and with a varying number of AMRs. Comparison against the commonly used Djikstra’s algorithm was also conducted (Shaikh and Dhale, 2013). Results show that the proposed A.N.T.s algorithm could execute 100 orders in a 27x23 layout with five robots 9.96 times faster than Dijkstra with no collisions. The algorithm is also shown to be able to help assign tasks to robots and help them find short paths to navigate their ways to the shelf units and packing stations. A.N.T.s could navigate traffic to avoid deadlocks and collisions in the warehouse with the aid of lanes and directions.

車用開門快速預警裝置

由於台灣汽、機車數量逐年攀升,且人口密度高、道路窄小和停車位不足,使得人、車爭道和兩車併排等現象層出不窮,導致汽車駕駛人或乘客在開啟車門時,未注意後方來車造成碰撞的傷亡事故屢屢發生。有鑑於開啟車門不當事故頻繁,本研究提出一個基於達靈頓電路(Darlington Circuit)的快速且成本低之車門開啟預警系統,當汽車駕駛人或乘客碰觸車門內把手內側的軟式觸控薄膜開關時,系統即會啟動欲開啟車門側的門外警示燈和車後方向燈光閃爍,以提早預警行人或機車騎士,同時,汽車A柱內的警示燈、蜂鳴器會同步閃爍和鳴響,以提醒汽車駕駛人或乘客注意後方車輛避免碰撞,保障用路人安全。

還有「轉環」的餘地—探討鋼球在自轉圓環軌道的運動

將鋼球放入一個鉛直站立的圓環凹槽內,當圓環由靜止開始轉動的過程中,鋼球沿著圓環向上運動,經過我們設計並改良的實驗架設,製作出穩定的實驗裝置進行量測,探討各種變因對鋼球運動的影響,並彙整出鋼球運動的幾個階段,進行理論建模與實驗數據的比較分析;我們參考幾篇論文嘗試在理論建模的方程式逐步加入模擬修正,從最單純的質點運動方程式逐步考慮旋轉、搖晃、阻力、鋼球體積等等,比較理論模擬結果與實驗數據的差異,也驗證了一些論文中缺乏實驗驗證的理論計算之正確性。

創新散熱系統-致冷晶片於電腦中的應用

1.手機及筆電已成不可或缺生活用品。但,若其充電器如果充電時間太久,可能會因為本身過熱而造成充電完畢要取回時,造成手部的燙傷。因為有此生活經驗,想改善充電器等高溫設備的散熱。 2.本研究探討電腦CPU上如果加上致冷晶片,如何加速散熱,增加使用安全外也進而提高電腦效能。 3.以電腦主機為實驗的操作器材,探究研發該散熱系統,並做為以後其他需要散熱的結構基礎,作為加速散熱的要件。目前致冷晶片在兩面溫差超越20°C時,可以使致冷晶片產生電能,接上微型馬達後,得以驅動,並產生對流散熱。 4.目前有關致冷晶片的實驗中,多數是使用致冷晶片和帕爾帖效應的關係,以通入電能使晶片兩面產生溫差。而本研究將採賽貝克效應做基礎,來探討新式散熱解決方案,使CPU產生的熱轉換成可以驅動微型馬達的電能。

以隨機噪音生成技術為基礎的驗證碼對抗式攻擊防禦機制

網路上常常會使用驗證碼(CAPTCHA)防止自動化程序取得網站資源,而一般而言,若驗證碼是可以輕易取得,十分容易被深度學習網路破解。然而,對抗式攻擊(adversarial attack)可以騙過許多深度學習網路。因此,本研究目的為建立能夠破解對抗式攻擊的深度學習網路。主要包含三個部分:建立Captcha breaker、使用對抗式攻擊影響breaker、防禦對抗式攻擊。Captcha breaker的部份使用模擬的目標驗證碼作為訓練資料,以解決訓練資料不足以及人工標籤的問題;而破解adversarial attack會使用adversarial training以及random noising的技術進行。

Art Recovery through PConv (Partial Convolutions) and GLCIC (Globally and Locally Consistent Image Completion)

在生成性模型(Generative Models)中的一個主要應用就是“影像修復” (Image Inpainting) 也稱為“影像完成”(Image completion)。 影像修復經常被應用於許多影像處理,包含在生活照片中移除背景不必要的物件再回填移除後缺損的影像。 但是,或許之前的研究較多著墨於技術而非美學,至目前為止,很少有影像修復的研究著重於藝術作品的重建應用。 所以,本研究計畫提出三個新的模型來針對藝術作品做更優化的影像修復,以達到較一般處理日常照片所使用的如Places2 和ImageNet等影像修復技術在視覺上更為自然逼真的處理: 第一種模型是PConv (Partial Convolutions),它利用部分旋積(partial convolution) 來避免一般由於遮蔽區域中畫素起始值設定而常見的影像模糊問題。 第二種模型是GLCIC (Globally and Locally Consistent Image Completion),是一種以GAN (Generative Adversarial Network) 為基礎,進一步在全域鑑別器 (global discriminator) 之上,再建構一個區域性鑑別器(local discriminator),以確保在整體畫面與細部畫面的合理與一致性的方法。 最後一個模型是一個在本研究中所提出的全新、整合性的模型–PConv-GAN。 在這個創新的模型中,我們將GLCIC模型中常用於旋積過程中”補零”(zero padding) 的手法,以PConv模型中部分旋積的方式來取代。最後我們會利用一系列的印象派畫作為例,以L1 loss 和PSNR (peak signal-to-noise ratio) 兩種方法來評估這三個模型。

Bio-Circular-Green Superabsorber

As the world has become concerned about the global waste crisis and global warming, there has been a surge of research within materials science to find materials that would replace plastic, such as bioplastics or biodegradable materials, in order to reduce environmental pollution. Plastics generates the microplastics that allowed them to become cross contamination enter the ocean through land, sea and river. Science research found (Lusher et al. (2017)) over 220 species of marine animals ingested microplastic, half of them are considered relevant for commercial purpose and increasing the risk of human consumption as it can induce immune response, oxidative stress, cytotoxicity, alter membrane integrity and cause differential expression of genes. Thailand is also experiencing such a challenge, as seen by the overabundance of plastic waste that might take centuries to decompose. For example, around 1680 million personal hygiene products such as diapers, sanitary napkins, and tampons are used each year. This study highlights the use of naturally accessible absorbent fibers from malva nut (Scaphium scaphigerum) (G. Don) Guib & Planch.), which is widely available and biodegradable in nature and has a low carbon footprint. This study also aimed to develop natural absorbent pads using compostable spun, external layers, and biodegradable glue. A prototype sanitary napkin with biodegradable absorbent pads was developed and evaluated for absorption ability, absorption rate, pH, and biodegradability. The absorbent material absorbed up to 19 times its weight in 2 minutes and 33 times its weight in 2 hours, which is enough for an average of 80-150 mL of menstrual blood. The prototype napkin deteriorated within 99 days, based on naked eye observation. Some signs of degradation and microorganisms growing on the prototype were also observed from scanning electron microscopic images. According to the findings, natural absorbent pads made from malva nut have the potential to be converted into sanitary napkins. Furthermore, it is proposed that the components, which include superabsorbent renewable materials, spinning compostable layer, external compostable layer and biodegradable glue, may be used in a variety of goods, including adult diaper pants, incontinence pads, and laboratory bench mats.