核心优势
- 最高性能
- 现代化设计
- 结构简单
FabImage Library Suite 是一款面向 C++ 和 .NET 程序员的机器视觉库。该软件包含一整套用于创建工业图像分析应用程序的函数，包括标准的图像采集接口、低级别图像处理例程以及多款即用型工具（如模板匹配、测量或条形码阅读器）。
该产品的主要优势包括强劲的性能、现代化的设计以及易于集成其他代码的简单结构。FabImage 库中可用的函数与 FabImage Studio 的过滤器紧密对应。
因此，可以在图形环境中快速设计算法的原型，然后将其转换为 C++或 .NET，甚至自动生成 C++ 代码。借助 Fabimage Library Suite，您可以立即获得最高质量、经过优化和经过现场测试的代码，以满足您的机器视觉项目的迫切需求！
Three types of Licenses are available:
Developer Licenses: Licenses required to develop a vision program.
Runtime Licenses: Licenses required to run a vision program. To purchase a Runtime License, you must have purchased a Developer License.
Add-on Licenses: Additional Licenses that allow you to expand the functionality of the other two types of Licenses.
Developer licenses
The Developer License is assigned to a single developer user and can be activated only via USB dongle. Free Technical Support services are included for the first 12 months after activation, such as:
- Most up-to-date version of the software with associated new features and documentation*
- Answers, via email, to technical questions related to the use of the software
After 12 months from the activation of the Developer License, it is necessary to purchase a Service License (FIL-EXT) in order to:
purchase additional Single Thread Runtime Licenses (FIL-RUN).
take advantage of the Technical Support services.
| * To obtain the most up-to-date version of the software, you must send Opto Engineering the WIBU file associated with the USB dongle of the License you wish to have the upgrade on. Learn more about how and where to download the WIBU file at https://docs.fab-image.com/stu... | |||
| P/N | Description | Category | Features |
| FIL-SUI | FabImage Library Suite (C++ and .NET) | Developer Basic License | Development environment (IDE) for direct programming in C++ or .NET. This type of License is suitable for those who prefer to program in C++ or C# and do not want to employ graphical programming. Allows multi-camera acquisition and development of processes (macrofilters) sequentially (single thread) |
| USB-DONGLE-FI | USB Dongle | Hardware | Required to activate the License via hardware USB dongle. |
ADD-ON License
ADD-ON Licenses** are additional Licenses that allow you to expand the functionality of the Basic License. To purchase ADD-ON Licenses, you must have purchased a Developer FabImage Library Suite License (FIL-SUI) .
| ** To order an ADD-ON License, you must also send the WIBU file associated with the USB dongle of the developer for which you wish to activate the add-on. Read more about how and where to download the WIBU file at this link. | |||
| *** It is not possible to build multiple macrofilters with Deep Learning that work in parallel. | |||
P/N |
Description |
Category |
Features |
FIL-PAR-ADD |
Parallel Processing ADD-ON for those who have purchased FabImage Library Suite (FIL-SUI) |
Developer ADD-ON License |
Allows the user to:
|
FI-DL-ADD |
FabImage Deep Learning ADD-ON for those who have already purchased FabImage Library Suite (FIL-SUI) |
Developer ADD-ON License |
Allows the user to use Deep Learning Tools ***. Visit the Deep Learning section for more information. |
Runtime Licenses
Runtime License SINGLE THREAD
The Runtime License is assigned to a single vision system and allows multi-camera acquisition and execution of processes (macrofilters) sequentially (single thread). It can be activated via two options:
USB Dongle (USB-DONGLE-RUN)
Computer ID*
To purchase a Single Thread Runtime License, you must have purchased the FabImage Library Suite Developer License (FIL-SUI). After 12 months from the activation of the Developer License, you are required to purchase the Service License (FIL-EXT) to continue purchasing Single Thread Runtime Licenses.
| * In case the License is lost through damage to the Computer, to which it is assigned by Computer ID, it cannot be recovered and a new one must be purchased. Opto Engineering recommends the option to purchase the license via USB dongle. | |||
P/N |
Description |
Category |
Features |
FIL-RUN |
FabImage Library Runtime |
Runtime License |
Allows you to run an unlimited number of processes (macrofilters) sequentially |
USB-DONGLE-RUN |
USB Dongle (Optional) |
Hardware |
The License is activated via USB dongle |
Runtime License MULTITHREADING
In order to run the Parallel Processing (FIL-PAR-ADD) features, you must purchase one of the following Runtime Licenses (these Runtime Libraries replace the FabImage Library Single Thread Runtime (FIL-RUN)). To purchase a Multithreading Runtime License, you must have purchased a FabImage Library Suite Developer License (FIL-SUI) and a Developer Parallel Processing ADD-ON License (FIL-PAR-ADD).
After 12 months from the activation of the Developer Parallel Processing ADD-ON License (FIL-PAR-ADD), you are required to purchase a Service License (FIL-EXT) if you wish to continue purchasing Multithreading Runtime Licenses.
P/N |
Description |
Category |
Features |
FIL-RUN-CL-XX* |
|||
FIL-RUN-CL-4 |
FabImage Library Runtime for a 4-core machine vision computer |
ADD-ON Runtime License for Parallel Processing |
Allows running an unlimited number of processes in parallel. Requires a PC with 4 cores. |
FIL-RUN-CL-6 |
FabImage Library Runtime for a 6-core machine vision computer |
ADD-ON Runtime License for Parallel Processing |
Allows running an unlimited number of processes in parallel. Requires a PC with 6 cores. |
FIL-RUN-CL-8 |
FabImage Library Runtime for an 8-core machine vision computer |
ADD-ON Runtime License for Parallel Processing |
Allows running an unlimited number of processes in parallel. Requires a PC with 8 cores. |
FIL-RUN-CL-16 |
FabImage Library Runtime for a 16-core machine vision computer |
ADD-ON Runtime License for Parallel Processing |
Allows running an unlimited number of processes in parallel. Requires a PC with 16 cores. |
FIL-RUN-TL-XX |
|||
FIL-RUN-TL-2 |
FabImage Library Runtime limited to 2 Threads |
ADD-ON Runtime License for Parallel Processing |
Enables the use of PCs with any number of Cores. The number of parallel processes is limited to 2 threads. |
FIL-RUN-TL-4 |
FabImage Library Runtime limited to 4 Threads |
ADD-ON Runtime License for Parallel Processing |
Enables the use of PCs with any number of Cores. The number of parallel processes is limited to 4 threads. |
FIL-RUN-TL-6 |
FabImage Library Runtime limited to 6 Threads |
ADD-ON Runtime License for Parallel Processing |
Enables the use of PCs with any number of Cores. The number of parallel processes is limited to 6 threads. |
FIL-RUN-TL-8 |
FabImage Library Runtime limited to 8 Threads |
ADD-ON Runtime License for Parallel Processing |
Enables the use of PCs with any number of Cores. The number of parallel processes is limited to 8 threads. |
FIL-RUN-TL-16 |
FabImage Library Runtime limited to 16 Threads |
ADD-ON Runtime License for Parallel Processing |
Enables the use of PCs with any number of Cores. The number of parallel processes is limited to 16 threads. |
* To run an unlimited number of processes in parallel, it is recommended to purchase the Runtime License, which corresponds to the number of cores on the machine vision computer.
ADD-ON Runtime License DEEP LEARNING
In order to run the Deep Learning ADD-ON features, the following Runtime License must be purchased in addition to the FabImage Library Runtime (FIL-RUN).
To purchase the Deep Learning ADD-ON Runtime, you must have purchased a FabImage Library Suite Developer License (FIL-SUI) and a Developer Deep Learning ADD-ON License (FI-DL-ADD). After 12 months from the activation of the Developer ADD-ON License (FI-DL-ADD), you will be required to purchase the Service License (DL-EXT), if you wish to purchase a Deep Learning ADD-ON Runtime License.
P/N | Description | Category | Features |
FIL-RUN-DL | FabImage Library Deep Learning ADD-ON Runtime | Deep Learning ADD-ON Runtime License. | Enables the user to execute single-threaded Deep learning* |
* Multiple GPU cards cannot be used for inference
After 12 months from the activation of the FabImage Library Suite Developer License (FIL-SUI) or Developer ADD-ON Licenses (FIL-PAR-ADD and FI-DL-ADD), it is required to purchase one of the following Service Licenses for:
Purchasing Additional Single-Threaded Runtime Licenses (FIL-RUN) or Multithreading Runtime Licenses (FIL-RUN-CL-xx or FIL-RUN-TL-xx)
Purchasing Additional Deep Learning ADD-ON Runtime Licenses (FIL-RUN-DL)
Maintaining active technical support
| * Parallel Add-on licenses get free updates once you purchase a FIL-EXT ** In order to purchase the DL-EXT, the FIL-EXT must also be purchased |
|||
P/N | Description | Category | Features |
FIL-EXT | FabImage Library Suite EXTENSION
| Service License | License required for:
3) Maintaining active technical support |
DL-EXT* | FabImage Deep Learning ADD-ON Extension | Service License | License required for: 1) Purchasing Additional Deep Learning ADD-ON Runtime Licenses 2) Maintaining active technical support |
Se non si estende il servizio di assistenza tecnica, è comunque possibile:
-utilizzare la licenza per sviluppatori (FIS-SUI) e le licenze add-on (FIL-PAR-ADD e FI-DL-ADD), ma solo nell'ultima versione disponibile alla scadenza del supporto tecnico.
-Acquistare e utilizzare le licenze Runtime (incluse quelle Multithreading e Deep Learning), ma solo nell'ultima versione disponibile alla scadenza del supporto tecnico.
IMPORTANTE
Opto Engineering rilascia da 2 a 4 nuove versioni delle licenze per sviluppatori ogni anno, quindi è molto probabile che alla scadenza del supporto sia già disponibile una nuova versione. Per questo motivo, oltre a poter usufruire del nostro supporto tecnico, Opto Engineering consiglia sempre l'acquisto di Licenze Service.
性能
FabImage Library Suite 将算法的精心设计与全方位的硬件优化相结合，形成出色的性能和速度，从而使得该库的性能和速度均跻身世界前列。我们使用多核处理器的 SSE 指令和并行计算进行各种操作。
现代化设计
各类数据均具有自动内存管理功能；可在异常情况下显式处理错误，且针对类型安全的特殊值使用可选类型。所有函数均为线程安全函数，且尽可能会在内部使用数据并行。
简易性与一致性
该库是一个简单的类型与函数集合，以具有适当标题的单个 DLL 文件的形式供用户使用。为实现最佳的可读性，在命名函数时应遵循一致的命名惯例（例如动词 + 名词形式，如下所示：SmoothImage、RotateVector）。所有结果均通过参考输出参数返回，所以十分有可能出现多个输出。
簡介
Deep Learning Add-on是机器视觉的一项突破性技术。它是一套由五种现成工具组成的软件，经由20-50个样本图像训练之后能够可对物体、缺陷或特征进行自动检测。内部使用了专为工业视觉系统设计和优化的大型神经网络。
与FabImage Studio Professional一起，您将获得一个用于训练和部署现代机器视觉应用的完整解决方案。
核心优势
從少數樣本中學習
一般的應用僅需要約20到50張影做訓練即可。愈多訓練的效果愈好，但我們的軟體根據訓練庫的主要參數屬性，內部將自動產生數以千計 新的人造影像，讓訓練更有效率。
運行在GPU和CPU
為了讓訓練更有效率，需要新一代的繪圖處理器(GPU)。在生產線上，您可以視需選用GPU或CPU，而GPU的處理時間會比CPU快3到10倍。(Object Classification的運行速度CPU和GPU一樣快)。
最佳性能
一般在GPU的訓練時間大約是5到15分鐘。推論時間的長短每張影像介於5到100毫秒，它和選用的工具與硬體有關。工業級的推論引擎 WEAVER可以保證最佳的性能。
訓練步驟
1. 收集影像並進行歸一化
- 取得20到50張影像，好壞都要，影像要盡可能涵蓋物件的變化，並儲存到硬碟裡。
- 影像中物件的比例、方向和明暗盡可能保持一致。
培训
- 打开FabImage Studio Professional并添加一个Deep Learning Add-on工具
- 打开与该工具关联的编辑器，并在那里加载培训图像
- 使用绘图工具标记图像或添加标记
- 点击“培训”
訓練庫和驗證庫
在深度學習中如同在機器學習的各個領域一樣，遵循正確的方法是非常重要的。最重要的規則是將訓練庫和驗證庫分開。 訓練庫是用於創建模型的一組樣本。我們不能用它來衡量模型的性能，因為這通常會產生過於樂觀的結果。因此，我們使用不同的數據(驗證庫)來評估模型。我們的深度學習工具會根據用戶提供的樣本自動創建這兩種資料庫。
執行
- 運行程式並查看結果
- 返回步驟1或2直到結果完全令人滿意。
特征检测
在監督模式下，用戶需要小心翼翼地根據訓練庫中的影像去標註缺陷的位置。該工具藉由尋找關鍵特徵去學習判斷好與壞。
太陽能板檢測
在這個應用中，必須在具有複雜特徵的背景上檢測出破損和刮傷。若用傳統的方法，將需要複雜的演算法，並針對每一種太陽能板調整數十種參數。使用深度學習，只需要一種工具就可以在監督模式下做訓練。
衛星影像識別
衛星影像不容易解析，因為它涵蓋非常多的特徵。不過，我們的深度學習軟體透過訓練可以非常穩定且可靠的區別道路和建築物。
異常檢測
非監督式模式的訓練更加簡單，沒有固定缺陷定義。我們的軟體將樣品圖像並進行訓練。此後尋找任何偏差。
包裝驗證
當壽司盒出貨到市場時，什麼位置放什麼食材都是固定的。雖然同一種食材的外觀不完全相同但可以接受， 而食材的缺陷卻不容易被定義。 解決方式是採用半監督模式去檢測到該工具在訓練期間所經歷的任何重大改變。
塑膠射出成型檢測
射出成型是一個複雜且伴隨許多生產問題的製程。塑膠件可能會折疊或其它形狀的變異，但對客戶來說都是可以接受的。 我們的深度學習工具可以從提供的樣本中接受這種變異，並在生產線上找出有任何異狀的不良品。
物件分類
物體分類工具根據物體的特徵將所提供的輸入圖影像為用戶特定的幾組。其結果為類別名稱和分類置信度。
瓶蓋：正面或反面
塑膠瓶蓋在生產時有時會不經意地翻面。假如客戶想要檢測這種情況，完全可以用傳統的視覺檢測方式做到， 只是它需要一位專家針對此應用設計特定的演算法。另一種方案，我們可以採用以深度學習為基礎的物件分類功能， 從訓練的影像庫中，自動學習瓶蓋的正反面。
3D鋁合金輪圈識別
一家工廠可能生產數以百計樣式不同的鋁合金輪圈。使用傳統的視覺檢測方式要在這麼多樣化的產品中識別出一個特定的輪圈， 實際上是不可能的。模板比對需要大量的時間嘗試從數百個模板中做匹配，而這些模板必須手工逐一訂製， 耗費太多開發和維護時間。而深度學習是一個相當理想的解決方案，它可以讓程式直接從影像庫中學習並訓練， 其結果非常的穩定且可靠。
個體區分
個體區分技術被用來在一張影像中定位、區隔和分類單一或多重物件。它不同於Feature Detection(特徵識別)技術，它可以找出個別的物件，即使物件互相接觸或是重疊也能夠正確區分。
堅果分類
混合堅果是一種非常受歡迎包含各種形狀堅果的休閒食品。由於包裝中堅果的成份百分比應和標籤上印刷的成分清單一致， 客戶希望確保每一種類型的正確數量都可以填充到包裝裡。Instance Segmentation (個體區分) 工具針對這種應用是非常理想的解決方案，因為它能區分物體跟物體間的範圍並返回被分割對象相對應的蒙版。
包裝驗證
歐洲常見的湯類蔬菜組合包是以隨機位置包裝在白色塑膠盤上。生產線作業員有時會忘記將某一種蔬菜放到盤子上。 雖然有秤重系統做把關，但客戶在包裝密封前還是想知道商品是否完整。由於不會有兩種蔬菜看起來是完全相同的， 最理想的解進方案是採用深度學習為基礎的分類法。在訓練階段，客戶只須要在所選的蔬菜上做標註即可。
特徵點定位
特徵點定位工具在影像中尋找獨特的形狀、特徵或記號做為定位點。和傳統的模板比對進行比較，這個工具會進行多次的樣本訓練， 避免因物件的變異影響定位的可靠度。
蜜蜂跟隨
使用我們的深度學習工具可以完成傳統影像處理方法似乎無法實現的任務。在這種情況下，我們可以使用深度學習來檢測蜜蜂， 當訓練完成時，我們可以檢查蜜蜂們是否感染了變色病-由寄生蟎侵襲蜜蜂所引起的疾病。寄生蟲會依附在他們的身體上， 並根據紅色斑點的發炎狀況來檢視他們是否健康。不會只有這個案例讓複雜的任務簡單化， 同時我們對於許多不同的產業是抱持著開放的態度，例如農業。
取放
在這些應用案例中，我們需要引導機器手臂抓取物品，最典型的是從輸送帶或容器中抓取物品。 有一個好的案例是抓取小的莖梗，然後將其垂直放置在花盆中。過程中檢測到任何不正確之處， 都有可能導致種植過深或上下顛倒，將導致根部無法順利生長。我們的深度學習工具可以快速定位植物的部位， 並提供準確的結果。
