小型光學震動分選站以高流量魚鬆中的細微魚刺為例

Abstract

魚鬆是我們日常中常見的食物用途廣泛，它經常被用在米飯、粥、包子、糕點和三明治等菜餚的配料或餡料上。它是由魚肉經過切塊、醃漬、烹飪、切絲、乾燥等等所製成。在這些處理之中還得處理一個問題:就是從一開始就與魚肉形影不離的魚刺，雖然不小心吃到魚刺是很常見的事情，也通常不會造成重大傷害。然而，根據魚刺的大小和卡住的位置，它可能會引起腸胃或是消化道的不適或輕傷，如果嚴重的話甚至有卡在喉嚨並導致窒息的風險，這些都會十分影響人購買以及食用魚鬆的意願。本篇論文從排除魚鬆中的魚刺這個前提結合了工業4.0(關於工業4.0在研究背景的部份會提到)的自動化概念:以設計出能夠在產線中使用的治具為目標從使產線作業人員能夠方便入料、接著透過震篩的概念、使用UV光來激發出魚鬆中魚刺的螢光再讓感測器與高速相機進行影像辨識、接著進行自動排除、到最後將沒有魚刺的魚鬆出料。期望整套系統能夠透過機構設計、軟體、硬體的相互配合、光學鏡頭辨識達到80%以上的魚刺辨識率與排除率，並真的能夠應用於產線上，使魚鬆的生產更為順利。

Fish floss is a common food in our daily life and has a wide range of uses. It is often used as ingredients or fillings in dishes such as rice, porridge, steamed buns, pastries and sandwiches. It is made from fish meat that has been diced, pickled, cooked, shredded, dried, etc. There is another problemto be dealt with in these treatments: the fishbone that has been inseparable from the fish from the beginning. Although it is very common to accidentally eat a fishbone, it usually does not cause major harm. However, depending on the size of the fishbone and where it gets stuck, it may cause discomfort or minor injury to the stomach or digestive tract. If it is serious, it may even get stuck in the throat and cause the risk of suffocation. These will greatly affect people's willingness to buy and eat fish floss. This thesis combines the automation concept of Industry 4.0 from the premise of removing the fishbone in the fish floss (industry 4.0 will be mentioned in the research background part): the goal is to design a fixture that can be used in the production line, from which the production line workers can conveniently feed materials, and then through the concept of vibrating sieve, use UV light to excite the fluorescence of the fishbone in the fish floss, andthen let the sensor and high-speed camera perform image recognition, then perform automatic removal, and finally discharge the fish floss without fishbone. It is hoped that the entire system can achieve a fishbone recognition rate and rejection rate of more than 80% through mechanism design, software and hardware cooperation, and optical lens recognition, and can really be applied to the production line to make the production of fish floss more smooth.