機器視覺檢測技術在工業自動化生產制造業帶來了新的改變！

　　信息獲取的75%以上來自于視覺，機器視覺是對于人類視覺的再提升。機器視覺檢測設備無疑比人眼更加強大，在很多環境下，人類的視覺變化已經越來越難滿足要求。例如電子制造等行業的高速高精產線，往往速度快，工件小，非標件多，如果大量依靠人工，工人必可避免的疲勞等客觀因素會帶來非常高的誤差，導致產品質量不穩定。因此，用機器視覺檢測設備來替代人眼至關重要。

　　More than 75% of information acquisition comes from vision, and machine vision is a further improvement of human vision. Machine vision detection equipment is undoubtedly more powerful than human eyes. In many environments, human visual changes have become increasingly difficult to meet the requirements. For example, high-speed and high-precision production lines in electronic manufacturing and other industries tend to be fast, with small workpieces and many non-standard parts. If a large number of workers rely on labor, objective factors such as fatigue that workers must avoid will bring very high errors, resulting in unstable product quality. Therefore, it is very important to replace human eyes with machine vision detection equipment.

　　隨著人工成本不可避免的不斷提高、工人不穩定性影響品質問題顯露，人的不確定因素成為勞動密集型制造的發展瓶頸。同時，我國產品制造模式的又一困境也在出現，即高端產品沒人做、做不出，而低端產品質量差、易積壓賣不出，產業結構非常不合理。

　　With the inevitable increase of labor costs and the exposure of the problem of workers' instability affecting quality, human uncertainty has become the bottleneck of the development of labor-intensive manufacturing. At the same time, another dilemma of China's product manufacturing model is also emerging, that is, no one can do high-end products, while low-end products are of poor quality, easy to overstock and cannot be sold, and the industrial structure is very unreasonable.

　　這些困境的根本出路就在于轉型升級，同時提升智能化程度，用人工智能結合制造業，以人工智能+機器人去減少人工，降低人工造成的品質不穩定，質量差等問題，無疑是促進制造業轉型升級的良好方法。

　　The fundamental way out of these difficulties lies in transformation and upgrading, and at the same time, improving the degree of intellectualization. Using artificial intelligence combined with manufacturing, using artificial intelligence + robots to reduce artificial, and reducing the problems of unstable quality and poor quality caused by artificial, is undoubtedly a good way to promote the transformation and upgrading of manufacturing.

　　機器視覺帶來的新改變，機器視覺是最基礎的單元技術，也是一種非接觸式感知裝置，在實際制造業的使用中，主要被用于自動獲取與解釋真實事物的影像，并取得圖像來進行機器識別，反饋并控制制造流程。

　　New changes brought about by machine vision. Machine vision is the most basic unit technology and a non-contact sensing device. In the actual manufacturing industry, it is mainly used to automatically obtain and interpret the image of real things, and obtain the image for machine recognition, feedback and control the manufacturing process.

　　當前機器視覺的典型實現技術方式是用工業相機來完成，工業相機通過集成運動控制單元，能把運動圖像采集到計算機，通過計算機識別反饋回到機器來執行動作調整。在原先，自動化設備是按照既有預定指令來執行，但加上視覺反饋后，同時也能根據對于環境感知，快速做出實時調整，使得設備智能性得到提升，執行的效率和準確性得到了提升。

　　At present, the typical implementation technology of machine vision is to use industrial camera. Through the integrated motion control unit, industrial camera can collect the moving image to the computer, and then return to the machine through computer recognition and feedback to perform action adjustment. In the past, the automatic equipment was implemented according to the existing predetermined instructions, but after adding visual feedback, it can also make real-time adjustments quickly according to the environmental perception, which improves the intelligence of the equipment and the efficiency and accuracy of implementation.

　　從生產線和智能工廠的角度來看，加入機器視覺后能很好提升制造水平。當前許多工廠設備已經實現了基本自動化，表現為能按照人的編程機械式重復運動，但如果給機器加上視覺，設備就能夠根據不同環境條件來進行主動調整，從而在自動化設備基礎上實現智能化。

　　From the perspective of production lines and intelligent factories, the addition of machine vision can well improve the manufacturing level. At present, many factory equipment have achieved basic automation, which is manifested in the mechanical repetitive movement according to human programming. However, if the machine is added with vision, the equipment can actively adjust according to different environmental conditions, so as to realize intellectualization on the basis of automatic equipment.

　　用機器視覺技術取代人工產業已經具備非常強的場景可行性。目前機器視覺在珠三角地區已經得到大范圍應用。由于長三角珠三角地區是電子制造產業集中地，加工電子元器件、Pcb板以及手機組裝等制造產業精細且比較小，同時有大量的強光作為背光，對人體的傷害非常大，這些場景無疑就非常適合機器視覺的應用。

　　It is very feasible to replace the artificial industry with machine vision technology. At present, machine vision has been widely used in the Pearl River Delta. As the Yangtze River Delta and Pearl River Delta are the concentration of electronic manufacturing industry, the manufacturing industries such as processing electronic components, PCB boards and mobile phone assembly are fine and relatively small, and there are a large number of strong lights as backlights, which do great harm to human body, these scenes are undoubtedly very suitable for the application of machine vision.

　　在未來，機器視覺的發展也有望大量替代人工，借助機器識別和信息處理，可以開拓的場景還有很多，機器視覺還可以提升工件的產品一致性和質量穩定性，機器視覺能很好完成檢測等制造關鍵流程，提升產線的整體效率。例如在檢測工件長度、寬度、厚度，以及例如在手機產業的屏幕粘乎和劃痕檢測等場景下，機器視覺都有著更多的發展空間。

　　In the future, the development of machine vision is also expected to replace labor in a large number. With the help of machine recognition and information processing, there are still many scenes that can be developed. Machine vision can also improve the product consistency and quality stability of workpieces. Machine vision can well complete key manufacturing processes such as inspection and improve the overall efficiency of the production line. For example, machine vision has more room for development in the detection of workpiece length, width and thickness, as well as in the detection of screen stickiness and scratches in the mobile phone industry.

　　除此之外，機器視覺還能實現更精準的定位，使得機器人在抓取過程中實現視覺定位和誤差補償。在 OCR識別上，機器視覺也大有可為，例如識別板面文字，判斷漏缺，判斷安裝是否正確。

　　In addition, machine vision can also achieve more accurate positioning, so that the robot can achieve visual positioning and error compensation in the process of grasping. In OCR recognition, machine vision also has great potential, such as recognizing board characters, judging omissions, and judging whether the installation is correct.

　　但單一的機器視覺的應用場景其實也非常有限，機器視覺發展到如今已經有10-20年歷史，在原先應用一直較少，隨著近年來攝像頭、光源等硬件成本的不斷降低，以及計算機處理能力的提升，人工智能技術的大量進入，才逐漸越來越廣泛進入到工業領域。

　　However, the application scenario of single machine vision is also very limited. Machine vision has developed for 10-20 years now, and it has been less used in the past. In recent years, with the continuous reduction of hardware costs such as cameras and light sources, as well as the improvement of computer processing capacity, artificial intelligence technology has entered the industrial field more and more widely.

　　軟件和智能技術的發展，使得機器視覺的應用明顯增多。在電子制造設備行業，機器視覺已經幾乎成為標配，但在具體應用上，卻往往還是采用傳統機器視覺算法，加入AI的機器視覺算法仍然應用較少，因此機器視覺+人工智能的領域其實還仍然有非常大的市場空間。

　　With the development of software and intelligent technology, the application of machine vision has increased significantly. In the electronic manufacturing equipment industry, machine vision has almost become a standard configuration, but in specific applications, traditional machine vision algorithms are often used, and machine vision algorithms with AI are still rarely used. Therefore, the field of machine vision + artificial intelligence still has a very large market space.

　　用人工智能+機器視覺來做機器識別，優勢非常明顯，最為直接的體現在于能大大增加制造柔性。

　　Using artificial intelligence + machine vision to do machine recognition has obvious advantages. The most direct embodiment is that it can greatly increase manufacturing flexibility.

　　柔性制造往往表現為機器人等工業設備能夠針對不同的位置、光照，以及復雜環境，呈現出更好的適應性。例如針對物體的顏色檢測，如果采用傳統的機器視覺，對于不同顏色的細微辨識度其實不高，某個顏色中如果出現深淺、偏色等輕微差別，傳統機器識別就很容易出錯，而用人工智能+機器視覺的方法就能夠比較準確提升檢測率和正確率。

　　Flexible manufacturing often shows that robots and other industrial equipment can show better adaptability to different positions, lighting, and complex environments. For example, for the color detection of objects, if traditional machine vision is used, the subtle recognition of different colors is not high. If there are slight differences such as depth and color deviation in a certain color, traditional machine recognition is easy to make mistakes, and the method of artificial intelligence + machine vision can be more accurate to improve the detection rate and accuracy.

　　采用AI的機器視覺來進行檢測，也能更加方便進行人機交互，提升檢測便利性。在原先，要實現精準點線面的指定位置檢測往往需要通過人進行點擊鼠標等操作，需要告訴視覺去檢測哪個圓哪條線，但如今加入AI后，電路板放到產線上，就能通過強化學習等方式自動識別原先各個方面的問題，從而完成視覺+智能標定、智能尺寸量測和位置實時判斷，效率和人機交互智能性都得到大大提升。

　　Using AI machine vision for detection can also facilitate human-computer interaction and improve the convenience of detection. In the past, to realize the specified position detection of precise points, lines and surfaces, people often needed to click the mouse and tell the vision to detect which circle and which line. But now, after adding AI, the circuit board is placed on the production line, and the problems in all aspects of the original can be automatically identified by means of reinforcement learning, so as to complete visual + intelligent calibration, intelligent dimension measurement and real-time position judgment, Efficiency and human-computer interaction intelligence have been greatly improved.