TWI750622B - Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium - Google Patents
Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium Download PDFInfo
- Publication number
- TWI750622B TWI750622B TW109111080A TW109111080A TWI750622B TW I750622 B TWI750622 B TW I750622B TW 109111080 A TW109111080 A TW 109111080A TW 109111080 A TW109111080 A TW 109111080A TW I750622 B TWI750622 B TW I750622B
- Authority
- TW
- Taiwan
- Prior art keywords
- test
- training
- value
- learning model
- deep learning
- Prior art date
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
Description
本案是關於人工智慧領域,特別是一種深度學習模型訓練系統及其方法。This case is about the field of artificial intelligence, especially a deep learning model training system and its method.
近年來人工智慧的研究日新月異,尤其是深度學習對於各個領域的應用更是突飛猛進,其中包括影像辨識領域。現今影像辨識領域的深度學習系統,通常需要使用者使用巨量的影像做為訓練資料來訓練深度學習系統,才能使深度學習系統得以正常運作,也就是使深度學習系統的輸出結果具有較高的置信度(又稱為信心度(Confidence))。雖然這樣的訓練方式在理論上是淺顯易懂的,但是在實行上卻並不容易,因為處理及獲得巨量的影像是相當耗時,通常需要高效能的處理器才能執行。獲得巨量的影像的方法除了蒐集原始影像之外,還有透過調整原始影像來獲得類似影像。In recent years, the research on artificial intelligence has changed rapidly, especially the application of deep learning in various fields, including the field of image recognition. Today's deep learning systems in the field of image recognition usually require users to use a huge amount of images as training data to train the deep learning system, so that the deep learning system can operate normally, that is, the output of the deep learning system has a higher level. Confidence (also known as Confidence). Although this training method is easy to understand in theory, it is not easy to implement because processing and obtaining a large number of images is time-consuming and usually requires a high-performance processor to execute. In addition to collecting the original images, the method to obtain a huge amount of images is to obtain similar images by adjusting the original images.
但是,以巨量的影像做為訓練資料來訓練深度學習系統的過程中,會發現訓練量跟置信度的增長並不成比例,例如大量的訓練量卻只讓置信度微幅提升。換句話說,現今透過調整原始影像來獲得類似影像的方法雖然可以增加訓練資料,但卻是一個耗時且並不有效的訓練方法。However, in the process of training a deep learning system with a huge amount of images as training data, it will be found that the training amount is not proportional to the increase in the confidence. For example, a large amount of training only increases the confidence slightly. In other words, the current method of obtaining similar images by adjusting the original image can increase the training data, but it is a time-consuming and inefficient training method.
另外,現今計量表的研究中,雖然已經研究出智慧計量表,智慧計量表透過電子式量測及回報資料,讓計量資料能即時上線。但是在實務上,智慧計量表並不普遍,大部分的使用者仍然是使用傳統計量表。若將傳統計量表更換為智慧計量表,對於自來水供應代表需要停水,而對於電力供應則需要停電,這樣不僅需耗費計量表本身的成本,對於使用者是24小時營運的工廠而言更需負擔龐大的時間成本,因此將傳統計量表更換為智慧計量表是使用者不樂見的。但是傳統計量表需要人工抄表來收集與統計計量資料,不僅沒效率又容易出錯,因此現今計量表仍需要改善。In addition, in the current meter research, although the smart meter has been researched, the smart meter can measure and report the data electronically, so that the measurement data can be online in real time. However, in practice, smart meters are not common, and most users still use traditional meters. If the traditional meter is replaced with a smart meter, the water supply needs to be cut off, but the power supply needs to be cut off, which not only costs the cost of the meter itself, but also requires a 24-hour operation for the user. It has a huge time cost, so users are not happy to replace the traditional meter with a smart meter. However, the traditional meter needs manual meter reading to collect and count the measurement data, which is not only inefficient but also prone to errors. Therefore, the current meter still needs to be improved.
鑑於上述,本案提出一種深度學習模型訓練系統及其方法。In view of the above, this case proposes a deep learning model training system and method.
依據一些實施例,深度學習模型訓練方法包括:依據多個測試值調整原始影像集以獲得多個測試影像集;輸入測試影像集至第一深度學習模型進行測試,獲得多個置信度,置信度以一對一的方式對應於測試值;依據各個測試值對應的置信度是否小於閥值,將測試值分類於有效組或無效組;依據無效組中的測試值,獲得多個訓練值;依據訓練值調整原始影像集以獲得多個訓練影像集進行訓練;以及,輸入原始影像集及訓練影像集至第一深度學習模型,獲得第二深度學習模型。According to some embodiments, the deep learning model training method includes: adjusting the original image set according to multiple test values to obtain multiple test image sets; inputting the test image set to the first deep learning model for testing to obtain multiple confidence levels, confidence levels Corresponding to the test values in a one-to-one manner; according to whether the confidence corresponding to each test value is less than the threshold, the test values are classified into valid groups or invalid groups; according to the test values in the invalid groups, multiple training values are obtained; The training value adjusts the original image set to obtain a plurality of training image sets for training; and, inputting the original image set and the training image set to the first deep learning model to obtain the second deep learning model.
依據一些實施例,深度學習模型訓練系統包括處理器及儲存裝置。處理器用於接收原始影像集、多個測試值及閥值。處理器用於依據測試值調整原始影像集以獲得多個測試影像集,並輸入測試影像集至第一深度學習模型進行測試而獲得多個置信度,置信度以一對一方式對應於測試值。處理器用依據各個測試值對應的置信度是否小於閥值將測試值分類於有效組或無效組,並依據無效組中的測試值獲得多個訓練值。處理器用於依據訓練值調整原始影像集以獲得多個訓練影像集,並且輸入原始影像集及訓練影像集至第一深度學習模型進行訓練而獲得第二深度學習模型。According to some embodiments, a deep learning model training system includes a processor and a storage device. The processor is used to receive the original image set, a plurality of test values and threshold values. The processor is configured to adjust the original image set according to the test value to obtain multiple test image sets, and input the test image set to the first deep learning model for testing to obtain multiple confidence levels, where the confidence levels correspond to the test values in a one-to-one manner. The processor classifies the test values into a valid group or an invalid group according to whether the corresponding confidence level of each test value is smaller than a threshold value, and obtains a plurality of training values according to the test values in the invalid group. The processor is configured to adjust the original image set according to the training value to obtain a plurality of training image sets, and input the original image set and the training image set to the first deep learning model for training to obtain the second deep learning model.
依據一些實施例,非暫態電腦可讀取儲存媒體用於儲存一或多個軟體程式,一或多個軟體程式包括多個指令,當這些指令被電子裝置的一或多個處理器執行時,將使電子裝置進行深度學習模型訓練方法。According to some embodiments, a non-transitory computer-readable storage medium is used to store one or more software programs, the one or more software programs including a plurality of instructions, when the instructions are executed by one or more processors of the electronic device , which will make the electronic device perform the deep learning model training method.
綜上,本案一些實施例的深度學習模型訓練系統、深度學習模型訓練系統方法及非暫態電腦可讀取儲存媒體,能依據測試值調整原始影像集以獲得測試影像集,並輸入測試影像集至第一深度學習模型進行測試而獲得測試影像集的置信度,也就是測試值的置信度。並且能依據閥值分類測試值為有效組或無效組,再從無效組中的測試值獲得訓練值。以及依據訓練值調整原始影像集以獲得訓練影像集,並輸入原始影像集及訓練影像集至第一深度學習模型進行訓練而獲得第二深度學習模型。To sum up, the deep learning model training system, the deep learning model training system method and the non-transitory computer-readable storage medium of some embodiments of the present application can adjust the original image set according to the test value to obtain the test image set, and input the test image set The first deep learning model is tested to obtain the confidence of the test image set, that is, the confidence of the test value. And the test value can be classified into the valid group or the invalid group according to the threshold value, and then the training value can be obtained from the test value in the invalid group. and adjusting the original image set according to the training value to obtain a training image set, and inputting the original image set and the training image set to the first deep learning model for training to obtain a second deep learning model.
以下將以圖式揭露本案之多個實施例,為明確說明起見,許多實務上的細節將在以下敘述中一併說明。然而,應瞭解到,這些實務上的細節不應用以限制本案。也就是說,在本案部分實施例中,這些實務上的細節是非必要的。此外,為簡化圖式起見,一些習知慣用的結構與元件在圖式中將以簡單示意的方式繪示之,而在所有圖式中,相同的標號將用於表示相同或相似的元件。且若實施上為可能,不同實施例的特徵係可以交互應用。Several embodiments of the present application will be disclosed in the following drawings. For the sake of clarity, many practical details will be described together in the following description. It should be understood, however, that these practical details should not be used to limit the present case. That is, in some embodiments of this case, these practical details are unnecessary. In addition, for the purpose of simplifying the drawings, some well-known and conventional structures and elements will be shown in a simple and schematic manner in the drawings, and the same reference numerals will be used to denote the same or similar elements in all the drawings. . And if possible in implementation, the features of different embodiments can be applied interactively.
圖1為根據本案一些實施例所繪示之深度學習模型訓練系統100的示意圖,請參照圖1,在一些實施例,深度學習模型訓練系統100包括處理器120及儲存裝置140。處理器120用於接收原始影像集、多個測試值及閥值,並依據測試值調整原始影像集以獲得多個測試影像集。處理器120用於輸入測試影像集至第一深度學習模型T1進行測試而獲得多個置信度,置信度以一對一方式對應於測試值。處理器120用於依據測試值對應的置信度是否小於閥值將測試值分類於有效組或無效組,並依據無效組中的測試值獲得多個訓練值。處理器120用於依據訓練值調整原始影像集以獲得多個訓練影像集。處理器120用於輸入原始影像集及訓練影像集至第一深度學習模型T1進行訓練而獲得第二深度學習模型T2。儲存裝置140用於儲存第一深度學習模型T1及第二深度學習模型T2。在一些實施例,儲存裝置140電性連接於處理器120。FIG. 1 is a schematic diagram of a deep learning
在一些實施例,深度學習模型(即,第一深度學習模型T1或第二深度學習模型T2)適於用在影像辨識,例如辨識影像中的數字圖像的數值。以具有數字圖像「0011」的影像為例,深度學習模型用於辨識影像而獲得數值「0011」。In some embodiments, the deep learning model (ie, the first deep learning model T1 or the second deep learning model T2 ) is suitable for use in image recognition, eg, recognizing numerical values of digital images in images. Taking an image with a digital image "0011" as an example, the deep learning model is used to identify the image to obtain the value "0011".
在一些實施例,當深度學習模型進行測試,深度學習模型依據輸入的測試問題(例如,具有數字圖像「0011」的影像)而輸出測試答案(例如,數值「0011」)及測試答案的置信度。具體而言,置信度代表深度學習模型對於測試答案是正確的有多少信心,或著說深度學習模型對於輸入的測試問題有多少信心能正確回答,也就是測試答案的正確率。因此,置信度介於0至1之間,當置信度為1代表測試答案的正確率是100%,置信度為0.5代表測試答案的正確率是50%,當置信度為0代表測試答案的正確率是0%。In some embodiments, when the deep learning model is tested, the deep learning model outputs the test answer (eg, the value "0011") and the confidence of the test answer according to the input test question (eg, the image with the digital image "0011") Spend. Specifically, the confidence represents how much confidence the deep learning model has for the test answer to be correct, or how much confidence the deep learning model has for the input test question to answer correctly, that is, the correct rate of the test answer. Therefore, the confidence level is between 0 and 1. When the confidence level is 1, it means that the correct rate of the test answer is 100%. When the confidence level is 0.5, it means that the correct rate of the test answer is 50%. When the confidence level is 0, it means that the test answer is correct. The correct rate is 0%.
在一些實施例,深度學習模型是以監督式學習進行訓練,具體而言,監督式學習是輸入相對應的訓練問題及訓練答案至深度學習模型進行訓練,於訓練之後的深度學習模型將可依據接收的訓練問題及訓練答案來處理測試問題,也就是深度學習模型能比對測試問題與訓練問題,並依據訓練答案而輸出對應測試問題的測試答案。因此,深度學習模型依據適當的監督式學習之後,深度學習模型能提高輸出的置信度,也就是測試答案的正確率。當本說明書提到相關於「輸入影像(集)至第一深度學習模型T1進行訓練,並獲得第二深度學習模型T2」時,代表「輸入影像(集)及對應的數值(集)至第一深度學習模型T1進行訓練,並獲得第二深度學習模型T2」。其中影像(集)是監督式學習中的訓練問題,對應的數值(集)是監督式學習中的訓練答案。對應的數值(集)不限於以任何形式獲得,例如影像(集)的檔案本身就具有數值(集)、從外部輸入數值(集)至深度學習模型訓練系統100、或經由處理器120依據其他辨識模型辨識影像(集)產生數值(集)。In some embodiments, the deep learning model is trained by supervised learning. Specifically, the supervised learning is inputting corresponding training questions and training answers to the deep learning model for training. After the training, the deep learning model can be trained according to The received training questions and training answers are used to process the test questions, that is, the deep learning model can compare the test questions and the training questions, and output the test answers corresponding to the test questions according to the training answers. Therefore, after the deep learning model is based on appropriate supervised learning, the deep learning model can improve the confidence of the output, that is, the correct rate of the test answer. When this specification refers to "input image (set) to the first deep learning model T1 for training, and obtain the second deep learning model T2", it means "input image (set) and corresponding values (set) to the first deep learning model T2" A deep learning model T1 is trained, and a second deep learning model T2 is obtained. The image (set) is the training question in supervised learning, and the corresponding value (set) is the training answer in supervised learning. The corresponding value (set) is not limited to be obtained in any form, for example, the file of the image (set) itself has the value (set), the value (set) is input to the deep learning
圖2為根據本案一些實施例所繪示之深度學習模型訓練方法的流程圖。為了清楚說明圖1之各項元件的運作,以下將搭配圖2之流程圖詳細說明如下。然而,本案所屬技術領域中具有通常知識者均可瞭解,本案實施例的深度學習模型訓練方法並不侷限應用於圖1的深度學習模型訓練系統100,也不侷限於圖2之流程圖的各項步驟順序。在一些實施例,非暫態電腦可讀取儲存媒體用於儲存一或多個軟體程式,一或多個軟體程式包括多個指令,當這些指令由電子裝置的一或多個處理電路執行時,將使電子裝置進行深度學習模型訓練方法。具體而言,電子裝置包括一或多個處理電路(又稱為控制電路)。深度學習模型訓練方法例如由一或多個軟體程式實作,軟體程式儲存於光碟、硬碟或其他非暫態電腦可讀取儲存媒體,軟體程式包括相關於處理電路的多個指令。當這些指令或軟體程式被電子裝置載入之後,將使電子裝置執行深度學習模型訓練方法。關於深度學習模型訓練方法的各項步驟的詳細說明,如下所列。FIG. 2 is a flowchart of a deep learning model training method according to some embodiments of the present application. In order to clearly illustrate the operations of the various elements in FIG. 1 , the following detailed description is given in conjunction with the flowchart in FIG. 2 . However, those with ordinary knowledge in the technical field to which this case belongs can understand that the deep learning model training method of the embodiment of this case is not limited to the deep learning
請一併參照圖1及圖2,在一些實施例,深度學習模型訓練方法的流程包括測試影像集獲得步驟(步驟S210)、模型測試步驟(步驟S220)、分組步驟(步驟S230)、訓練值獲得步驟(步驟S240)、訓練影像集獲得步驟(步驟S250)及模型訓練步驟(步驟S260)。Please refer to FIG. 1 and FIG. 2 together. In some embodiments, the process of the deep learning model training method includes a test image set acquisition step (step S210 ), a model testing step (step S220 ), a grouping step (step S230 ), a training value The obtaining step (step S240 ), the training image set obtaining step (step S250 ), and the model training step (step S260 ).
在一些實施例,測試影像集獲得步驟(圖2之步驟S210)包括:依據多個測試值調整原始影像集以獲得多個測試影像集。具體而言,原始影像集包括多個原始影像,每個測試影像集分別包括多個測試影像。測試值以角度旋轉為例,測試值的單位為「度」,各個測試值分別為「-10、-9.5、-9、-8.5、-8、-7.5、-7、-6.5、-6、-5.5、-5、-4.5、-4、-3.5、-3、-2.5、-2、-1.5、-1、-0.5、0、0.5、1、1.5、2、2.5、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」。測試值為正值表示將原始影像朝向順時針旋轉,測試值為負值表示將原始影像朝向逆時針旋轉,測試值為「0」則表示未旋轉原始影像,但並不以此為限。在一些實施例,測試影像集以一對一的方式對應於測試值,例如旋轉「-10」度為一個測試影像集,旋轉「-9.5」度為另一個測試影像集,以此類推,因此當測試值有「21」個時,處理器120就依據「21」個測試值調整原始影像集以獲得對應的「21」個測試影像集。在一些實施例,一個原始影像集之中原始影像的數目乘以測試比例係數等於一個測試影像集之中測試影像的數量,測試比例係數介於0至1之間。具體而言,處理器120依據測試比例係數從原始影像集之中選出部分的原始影像集,並且調整部分的原始影像集以獲得各個測試影像集。需特別說明的是,「部分的原始影像集之中的原始影像的數目」(或是,「一個測試影像集之中測試影像的數量」)除以「原始影像集之中的原始影像的數目」等於測試比例係數。例如原始影像的數目為「1000」、測試值共有「21」個及測試比例係數為「0.1」時,部分的原始影像的數目為「1000*0.1=100」,各個測試影像集之中測試影像的數量為「100」,因此「21」個測試影像集之中測試影像的總數為「21*100=2100」。In some embodiments, the step of obtaining a test image set (step S210 in FIG. 2 ) includes: adjusting the original image set according to a plurality of test values to obtain a plurality of test image sets. Specifically, the original image set includes a plurality of original images, and each test image set includes a plurality of test images respectively. The test value takes angle rotation as an example, the unit of test value is "degree", and each test value is "-10, -9.5, -9, -8.5, -8, -7.5, -7, -6.5, -6, -5.5, -5, -4.5, -4, -3.5, -3, -2.5, -2, -1.5, -1, -0.5, 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4 , 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10”. A positive test value indicates that the original image is rotated clockwise, a negative value indicates that the original image is rotated counterclockwise, and a test value of "0" indicates that the original image is not rotated, but not limited to this. In some embodiments, test image sets correspond to test values in a one-to-one manner, eg rotated "-10" degrees for one test image set, "-9.5" degrees for another test image set, and so on, so When there are "21" test values, the
在一些實施例,測試值為等差數列,例如前述角度旋轉的實施例,測試值的最大值為「10」度,測試值的最小值為「-10」度,呈等差數列的測試值之公差為「0.5」。In some embodiments, the test value is an arithmetic sequence. For example, in the above-mentioned angular rotation embodiment, the maximum value of the test value is "10" degrees, and the minimum value of the test value is "-10" degrees. The tolerance is "0.5".
在一些實施例,當處理器120依據測試值調整原始影像集以獲得多個測試影像集時,各個測試值只對應同一個調整類別,例如調整類別為角度旋轉,但是不以此為限。在一些實施例,調整類別例如但不限於角度旋轉、等比例縮放、高度放大、寬度放大、對比度調整、亮度調整及伽瑪(Gamma)值調整等。其中「角度旋轉」能以任意方向旋轉原始影像,例如以直角坐標系為例,原始影像包括第一軸向及第二軸向,其中第一軸向為Y軸,第二軸向為Z軸。原始影像能以X軸、Y軸、Z軸或任意軸向旋轉為測試影像。「高度放大」為原始影像在維持等寬的條件下拉長,「寬度放大」為原始影像在維持等高的條件下拉寬。In some embodiments, when the
在一些實施例,測試值包括原始影像集對應的原始值。具體而言,原始值為測試值的其中之一,通常原始值為測試值的基準,例如原始值為測試值之中的最小值、最大值或中位數。以測試值對應的調整類別而言,原始值為原始影像集對應於調整類別的數值。例如調整類別為角度旋轉時,原始值為「0」度,也就是原始影像集的旋轉角度為「0」度(即,未旋轉)。例如調整類別為等比例縮放時,原始值為「1」,也就是原始影像集的比例為「1」倍(即,原始尺寸)。In some embodiments, the test values include raw values corresponding to the original image set. Specifically, the original value is one of the test values, usually the original value is a benchmark of the test value, for example, the original value is the minimum value, the maximum value or the median among the test values. In terms of the adjustment category corresponding to the test value, the original value is the value corresponding to the adjustment category of the original image set. For example, when the adjustment type is angle rotation, the original value is "0" degrees, that is, the rotation angle of the original image set is "0" degrees (ie, not rotated). For example, when the adjustment category is proportional scaling, the original value is "1", that is, the scale of the original image set is "1" times (ie, the original size).
在一些實施例,原始值對應的測試影像集為部分的原始影像集。具體而言,當處理器120依據原始值調整原始影像集以獲得測試影像集時,處理器120以原始影像集做為測試影像集,也就是處理器120無需調整原始影像集就可獲得測試影像集。例如測試值對應的調整類別為角度旋轉時,原始值為「0」度,因此處理器120能將原始影像集做為測試影像集。在一些實施例,處理器120依據測試比例係數決定部分的原始影像集為測試影像集。In some embodiments, the test image set corresponding to the original value is a partial original image set. Specifically, when the
在一些實施例,模型測試步驟(步驟S220)包括:輸入測試影像集至第一深度學習模型T1進行測試,獲得多個置信度,置信度以一對一的方式對應於測試值。由前述關於深度學習模型的說明可知,置信度代表第一深度學習模型T1對於輸入的測試影像集有多少信心能正確輸出對應的測試答案。因此,當處理器120輸入測試影像集至第一深度學習模型T1進行測試時,處理器120能從第一深度學習模型T1獲得測試影像集對應的置信度。由於測試影像集是依據測試值所調整,置信度以一對一的方式對應於測試值。在一些實施例,測試影像集對應的置信度為多個子置信度的平均數。具體而言,測試影像集包括多個測試影像,第一深度學習模型T1依據輸入的測試影像集而對應輸出置信度,並且第一深度學習模型T1依據輸入的各個測試影像而對應輸出各個子置信度。因此處理器120輸入任一個測試影像至第一深度學習模型T1進行測試時,第一深度學習模型T1能輸出對應於這個測試影像的子置信度,因此處理器120將某一個測試影像集之中的各個測試影像所對應的所有子置信度平均計算,能獲得這個測試影像集的置信度。在一些實施例,測試影像集對應的置信度可以為多個子置信度的中位數、最大值、最小值,不以此為限。In some embodiments, the model testing step (step S220 ) includes: inputting a test image set to the first deep learning model T1 for testing, and obtaining a plurality of confidence levels, where the confidence levels correspond to the test values in a one-to-one manner. As can be seen from the foregoing description about the deep learning model, the confidence level represents how much confidence the first deep learning model T1 has in the input test image set to correctly output the corresponding test answer. Therefore, when the
在一些實施例,測試值以角度旋轉為例,測試值與置信度的對照表如下表1所示:In some embodiments, the test value takes the angle rotation as an example, and the comparison table between the test value and the confidence is shown in Table 1 below:
表1
在一些實施例,分組步驟(圖2之步驟S230)包括:判斷各個測試值對應的置信度是否小於閥值,將測試值分類於無效組或有效組。具體而言,置信度越高代表第一深度學習模型T1越能正確辨識輸入之測試影像集,置信度越低代表第一深度學習模型T1越不能正確辨識。因此處理器120判斷測試值對應的置信度是否小於閥值,透過閥值區分有效的置信度及無效的置信度,將置信度小於閥值的測試值分類於無效組,並且將置信度大於或等於閥值的測試值分類於有效組。以測試值為角度旋轉為例並參照表1,假如閥值為「0.5」,當置信度小於「0.5」代表第一深度學習模型T1不能有效的正確辨識,因此置信度小於「0.5」對應的測試值「-10、-9.5、-9、-8.5、-8、-7.5、-7、-6.5、-6、-5.5、-5、-4.5、-4、-3.5、-3、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」分類於無效組。當置信度大於或等於「0.5」代表第一深度學習模型T1能有效的正確辨識,因此置信度大於或等於「0.5」對應的測試值「-2.5、-2、-1.5、-1、-0.5、0、0.5、1、1.5、2、2.5」分類於有效組。In some embodiments, the grouping step (step S230 in FIG. 2 ) includes: judging whether the confidence level corresponding to each test value is less than a threshold value, and classifying the test values into invalid groups or valid groups. Specifically, a higher confidence level means that the first deep learning model T1 can more correctly identify the input test image set, and a lower confidence level means that the first deep learning model T1 cannot be correctly identified. Therefore, the
在一些實施例,測試值以角度旋轉為例,無效組與有效組的對照表如下表2所示:In some embodiments, the test value takes the angle rotation as an example, and the comparison table between the invalid group and the valid group is shown in Table 2 below:
表2
在一些實施例,訓練值獲得步驟(圖2之步驟S240)包括:依據無效組中的測試值,獲得多個訓練值。具體而言,對應於無效組中的測試值的測試影像集是第一深度學習模型T1無法有效的正確辨識的測試影像集,因此對應於無效組中的測試值的測試影像集正是第一深度學習模型T1需要加強訓練的部分,所以處理器120依據無效組中的測試值而獲得訓練值。在一些實施例,訓練值為無效組中的所有測試值,以測試值為角度旋轉為例及參照表2,訓練值為「-10、-9.5、-9、-8.5、-8、-7.5、-7、-6.5、-6、-5.5、-5、-4.5、-4、-3.5、-3、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」。In some embodiments, the step of obtaining the training value (step S240 in FIG. 2 ) includes: obtaining a plurality of training values according to the test values in the invalid group. Specifically, the test image set corresponding to the test values in the invalid group is the test image set that cannot be effectively and correctly identified by the first deep learning model T1, so the test image set corresponding to the test values in the invalid group is the first The deep learning model T1 needs to strengthen the training part, so the
在一些實施例,訓練值獲得步驟(圖2之步驟S240)更包括:依據有效組中最大的測試值與原始值之間的差值,獲得有效區間值;以及依據有效區間值的多個整數倍,從無效組中的測試值獲得訓練值。具體而言,處理器120依據無效組中的測試值獲得訓練值的過程包括上述步驟,其中處理器120運算有效組中最大的測試值與原始值之間的差值,再由差值的兩倍獲得有效區間值(即,有效區間值是有效組中最大的測試值與原始值之間的兩倍差值),並且處理器120將無效組中符合有效區間值的多個整數倍的測試值做為訓練值。以測試值為角度旋轉為例及參照表2,有效組中最大的測試值為「2.5」,原始值為「0」,有效組中最大的測試值與原始值之間的差值為「|2.5-0|=2.5」,差值的兩倍為「2*2.5=5」,因此有效區間值為「5」。有效區間值的多個整數倍為「5*m,m=0、±1、±2、…」,因此無效組中符合有效區間值的多個整數倍的測試值為「-10、-5、5、10」,所以訓練值為「-10、-5、5、10」。In some embodiments, the step of obtaining the training value (step S240 in FIG. 2 ) further includes: obtaining a valid interval value according to the difference between the largest test value in the valid group and the original value; and according to a plurality of integers of the valid interval value times, the training values are obtained from the test values in the null group. Specifically, the process of obtaining the training value by the
在一些實施例,訓練影像集獲得步驟(圖2之步驟S250)包括:依據訓練值調整原始影像集以獲得多個訓練影像集。具體而言,訓練影像集包括多個訓練影像。以測試值為角度旋轉為例並參照表2,訓練值的單位為「度」,各個訓練值為「-10、-5、5、10」。在一些實施例,訓練影像集以一對一的方式對應於訓練值,例如將原始影像集旋轉「-10」度以獲得一個訓練影像集,將原始影像集旋轉「-5」度以獲得另一個訓練影像集,以此類推,因此當訓練值有「4」個時,處理器120就依據「4」個訓練值調整原始影像集以獲得「4」個訓練影像集。在一些實施例,一個原始影像集之中原始影像的數目乘以訓練比例係數等於一個訓練影像集之中訓練影像的數量,訓練比例係數介於0至1之間。具體而言,訓練比例係數類似於測試比例係數,差異在於訓練比例係數例如但不限於等於測試比例係數。處理器120依據訓練比例係數從原始影像集之中選出部分的原始影像集,並且調整部分的原始影像集以獲得各個訓練影像集。需特別說明的是,「部分的原始影像集之中的原始影像的數目」(或是,「一個訓練影像集之中訓練影像的數量」)除以「原始影像集之中的原始影像的數目」等於訓練比例係數。例如原始影像的數目為「1000」及訓練比例係數為「0.5」時,一個訓練影像集之中訓練影像的數量為「1000*0.5=500」,同理對於「4」個訓練影像集之中訓練影像的總數為「4*500=2000」。In some embodiments, the step of obtaining the training image set (step S250 in FIG. 2 ) includes: adjusting the original image set according to the training value to obtain a plurality of training image sets. Specifically, the training image set includes a plurality of training images. Taking the test value as an example of an angle rotation and referring to Table 2, the unit of the training value is "degree", and each training value is "-10, -5, 5, 10". In some embodiments, training image sets correspond to training values in a one-to-one manner, eg, rotating the original image set by "-10" degrees to obtain one training image set, and rotating the original image set by "-5" degrees to obtain another One training image set, and so on, so when there are "4" training values, the
在一些實施例,模型訓練步驟(圖2之步驟S260)包括:輸入原始影像集及訓練影像集至第一深度學習模型T1進行訓練,獲得第二深度學習模型T2。具體而言,當處理器120輸入原始影像集及訓練影像集至第一深度學習模型T1進行訓練時,第一深度學習模型T1以原始影像集及各個訓練影像集做為訓練資料。換句話說,第二深度學習模型T2能依據原始影像集及各個訓練影像集來提高無效組的各個測試影像集的置信度。因此,相對於第一深度學習模型T1,在不需要以無效組中的全部測試值做為訓練值的情況下,第二深度學習模型T2能依據各個訓練影像集做為訓練資料,提高無效組的各個測試影像集所對應的置信度,因此能精簡深度學習模型的訓練資料量。以測試值為角度旋轉為例並參照表2,處理器120只需要輸入原始影像集及訓練值為「-10、-5、5、10」對應的訓練影像集(總共4個訓練影像集)至第一深度學習模型T1進行訓練。In some embodiments, the model training step (step S260 in FIG. 2 ) includes: inputting the original image set and the training image set to the first deep learning model T1 for training to obtain the second deep learning model T2 . Specifically, when the
在一些實施例,以訓練比例係數是1為例,訓練影像集之訓練影像的數目等於原始影像集之中原始影像的數目,對於以無效組中的全部測試值(即,-10、-9.5、-9、-8.5、-8、-7.5、-7、-6.5、-6、-5.5、-5、-4.5、-4、-3.5、-3、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10)做為訓練值(總共30個訓練影像集,其中不包含原始影像集)的情況做比較,深度學習模型訓練系統100及深度學習模型訓練方法只需「(1+4)/(1+30)=16.1%」的訓練資料量。對於以全部測試值做為訓練值(總共41個訓練影像集,其中包含原始影像集)的情況做比較,深度學習模型訓練系統100及深度學習模型訓練方法只需「(1+4)/(1+40)=12.2%」的訓練資料量。上述精簡訓練資料量的實施例僅做為範例並不以此為限,學習模型訓練系統100及深度學習模型訓練方法所精簡的訓練資料量將依據不同的調整類別、不同的測試值、不同的置信度及不同的閥值而有不同的結果。In some embodiments, taking the training scale factor of 1 as an example, the number of training images in the training image set is equal to the number of original images in the original image set, for all test values in the null set (ie, -10, -9.5 , -9, -8.5, -8, -7.5, -7, -6.5, -6, -5.5, -5, -4.5, -4, -3.5, -3, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10) as training values (30 training image sets in total, excluding the original image set) for comparison, the deep learning
在一些實施例,測試值以角度旋轉為例,測試值與置信度的對照表如下表3所示:In some embodiments, the test value takes the angle rotation as an example, and the comparison table between the test value and the confidence is shown in Table 3 below:
表3
在一些實施例,依據表3各個測試值分別為「0、0.5、1、1.5、2、2.5、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」,原始值為「0」度。由於閥值為「0.5」,因此置信度小於「0.5」的測試值「3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」分類於無效組,置信度大於或等於「0.5」的測試值「0、0.5、1、1.5、2、2.5」分類於有效組。在一些實施例,訓練值為無效組中的所有測試值時,訓練值為「3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」。在一些實施例,訓練值獲得步驟(圖2之步驟S250)更包括:依據有效組中最大的測試值與原始值之間的差值,獲得有效區間值;以及依據有效區間值的多個整數倍,從無效組中的測試值獲得訓練值。有效組中最大的測試值為「2.5」,原始值為「0」,有效組中最大的測試值與原始值之間的差值為「|2.5-0|=2.5」,差值的兩倍為「2*2.5=5」,因此有效區間值為「5」。有效區間值的多個整數倍為「5*m,m=0、±1、±2、…」,因此無效組中符合有效區間值的多個整數倍的測試值為「5、10」,所以訓練值為「5、10」。In some embodiments, each test value according to Table 3 is "0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10", the original value is "0" degrees. Since the threshold is "0.5", the test values "3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10" with confidence less than "0.5" are classified In the invalid group, test values "0, 0.5, 1, 1.5, 2, 2.5" with confidence greater than or equal to "0.5" are classified as valid. In some embodiments, the training value is "3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10 when the training value is all test values in the null group. ". In some embodiments, the step of obtaining the training value (step S250 in FIG. 2 ) further includes: obtaining a valid interval value according to the difference between the largest test value in the valid set and the original value; and according to a plurality of integers of the valid interval value times, the training values are obtained from the test values in the null group. The largest test value in the valid group is "2.5", the original value is "0", and the difference between the largest test value in the valid group and the original value is "|2.5-0|=2.5", twice the difference is "2*2.5=5", so the valid interval value is "5". Multiple integer multiples of the valid interval value are "5*m, m=0, ±1, ±2, ...", so the test values of multiple integer multiples of the valid interval value in the invalid group are "5, 10", So the training value is "5, 10".
在一些實施例,深度學習模型訓練方法更包括重要度獲得步驟,重要度獲得步驟用於獲得測試值對應的重要度。在一些實施例,處理器120依據重要度獲得步驟獲得測試值對應的重要度。具體而言,當處理器120依據測試值調整原始影像集為多個測試影像集時,各個測試值只對應同一個調整類別,因此重要度獲得步驟用於獲得調整類別的重要度。因此調整類別以一對一的方式對應於重要度。In some embodiments, the deep learning model training method further includes an importance obtaining step, and the importance obtaining step is used to obtain the importance corresponding to the test value. In some embodiments, the
在一些實施例,深度學習模型訓練系統100及深度學習模型訓練方法能依據各個調整類別的重要度的大小,決定以那個調整類別對第一深度學習模型T1進行訓練。例如,以重要度最小的角度旋轉做為調整類別。在一些實施例,深度學習模型訓練系統100及深度學習模型訓練方法能依據各個調整類別的重要度的大小,依序以各個調整類別對第一深度學習模型T1進行訓練。In some embodiments, the deep learning
在一些實施例,原始影像的高大於原始影像的寬時,調整類別的重要度的大小排列如下所列:等比例縮放>角度旋轉>對比度調整>亮度調整>伽瑪值調整>高度放大>寬度放大。In some embodiments, when the height of the original image is greater than the width of the original image, the importance of the adjustment categories is arranged as follows: proportional scaling > angle rotation > contrast adjustment > brightness adjustment > gamma value adjustment > height enlargement > width enlarge.
在一些實施例,原始影像的高小於原始影像的寬時,調整類別的重要度的大小排列如下所列:等比例縮放>角度旋轉>對比度調整>亮度調整>伽瑪值調整>寬度放大>高度放大。In some embodiments, when the height of the original image is smaller than the width of the original image, the importance of the adjustment categories is arranged as follows: scaling>angle rotation>contrast adjustment>brightness adjustment>gamma value adjustment>width enlargement>height enlarge.
圖3為根據本案一些實施例所繪示之重要度獲得步驟的流程圖。請參照圖3,在一些實施例,重要度獲得步驟的流程包括範圍值獲得步驟(步驟S310)、分組步驟(步驟S320)、無效邊界值獲得步驟(步驟S330)及重要度決定步驟(步驟S340)。FIG. 3 is a flow chart of the steps of obtaining the importance according to some embodiments of the present application. Referring to FIG. 3 , in some embodiments, the flow of the importance obtaining step includes a range value obtaining step (step S310 ), a grouping step (step S320 ), an invalid boundary value obtaining step (step S330 ), and an importance determining step (step S340 ) ).
在一些實施例,範圍值獲得步驟(圖3之步驟S310)包括:依據最大的測試值與最小的測試值之間的差值,獲得範圍值。具體而言,範圍值為測試值的範圍,因此處理器120能依據最大的測試值與最小的測試值之間的差值獲得範圍值。以測試值為角度旋轉為例及參照表1,最大的測試值為「10」,最小的測試值為「-10」,因此範圍值為「|10-(-10)|=20」。In some embodiments, the step of obtaining the range value (step S310 in FIG. 3 ) includes: obtaining the range value according to the difference between the largest test value and the smallest test value. Specifically, the range value is the range of the test value, so the
在一些實施例,分組步驟(圖3之步驟S320)包括:依據各個測試值對應的置信度是否小於閥值,將測試值分類於無效組或有效組。分組步驟(圖3之步驟S320)類似於分組步驟(圖2之步驟S230),於此不再贅述,步驟S320與步驟S230之間的差異在於兩者的閥值可以相同(例如,閥值都是「0.5」),也可以不相同(例如步驟S320的閥值可以是「0.6」,步驟S230的閥值可以是「0.5」)。以測試值為角度旋轉為例及參照表2,假如閥值為「0.5」,置信度小於「0.5」而分類於無效組的測試值為「-10、-9.5、-9、-8.5、-8、-7.5、-7、-6.5、-6、-5.5、-5、-4.5、-4、-3.5、-3、3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」。當置信度大於或等於「0.5」而分類於有效組的測試值為「-2.5、-2、-1.5、-1、-0.5、0、0.5、1、1.5、2、2.5」。In some embodiments, the grouping step (step S320 in FIG. 3 ) includes: classifying the test values into invalid groups or valid groups according to whether the confidence levels corresponding to each test value are smaller than a threshold. The grouping step (step S320 in FIG. 3 ) is similar to the grouping step (step S230 in FIG. 2 ), which will not be repeated here. The difference between step S320 and step S230 is that the thresholds of both may be the same (for example, is "0.5"), or may be different (for example, the threshold value in step S320 may be "0.6", and the threshold value in step S230 may be "0.5"). Taking the test value of angle rotation as an example and referring to Table 2, if the threshold value is "0.5", the confidence level is less than "0.5" and the test value classified in the invalid group is "-10, -9.5, -9, -8.5, - 8, -7.5, -7, -6.5, -6, -5.5, -5, -4.5, -4, -3.5, -3, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10”. When the confidence level is greater than or equal to "0.5", the test value classified as a valid group is "-2.5, -2, -1.5, -1, -0.5, 0, 0.5, 1, 1.5, 2, 2.5".
在一些實施例,無效邊界值獲得步驟(圖3之步驟S330)包括:依據無效組中的測試值與原始值之間最小的差值,獲得無效邊界值。具體而言,處理器120能比較無效組中的測試值與原始值之間的差距,獲得無效組中的各個測試值與原始值之間分別對應的各個差值,並且透過比較這些差值的大小從中獲得最小的差值,再依據最小的差值對應的測試值獲得無效邊界值,也就是無效邊界值為最小差值對應的測試值。需特別說明的是,測試值與原始值之間的差值為「|原始值-測試值|」,也就是各個差值皆為正值。其中無效邊界值代表測試值無法被第一深度學習模組T0正確辨識的門檻,換句話說,無效邊界值是無效組與有效組之間的分界,並且無效邊界值為無效組中的測試值。在一些實施例,以測試值為角度旋轉為例及參照表2,假如原始值為「0」,無效組中的測試值與原始值之間的差值為「3、3.5、4、4.5、5、5.5、6、6.5、7、7.5、8、8.5、9、9.5、10」,因此最小的差值為「3」,而最小的差值對應的測試值為「3」或「-3」,並且「3」或「-3」為無效邊界值。需特別說明的是,以「3」或「-3」做為無效邊界值並不影響重要度,因為「3」及「-3」與原始值之間的差值相同(皆為最小的差值「3」)。In some embodiments, the step of obtaining the invalid boundary value (step S330 in FIG. 3 ) includes: obtaining the invalid boundary value according to the smallest difference between the test value in the invalid group and the original value. Specifically, the
在一些實施例,重要度決定步驟(圖3之步驟S340)包括:依據原始值與無效邊界值之間的差值除以範圍值,獲得重要度。具體而言,處理器120能依據原始值與無效邊界值之間的差值除以範圍值獲得商值,而商值即為重要度。原始值為原始影像集對應於調整類別的數值,重要度為原始值與無效邊界值之間的差值除以範圍值,也就是重要度對應於有效組(無效組)與範圍值之間的比例。因此,當重要度越低,代表有效組的測試值佔所有的測試值的比例越低(無效組的測試值佔所有的測試值的比例越高),也就是這個調整類別越容易影響置信度。反之,當重要度越高,代表有效組的測試值佔所有的測試值的比例越高(無效組的測試值佔所有的測試值的比例越低),也就是這個調整類別越不容易影響置信度。以測試值為角度旋轉為例及參照表2,以測試值為角度旋轉為例及參照表2,原始值是「0」,無效邊界值與原始值之間的差值是「3」,範圍值是「20」,商值是「3/20=0.15」,因此重要度是「0.15」。In some embodiments, the step of determining the importance level (step S340 in FIG. 3 ) includes: dividing the difference between the original value and the invalid boundary value by the range value to obtain the importance level. Specifically, the
在一些實施例(例如表2),重要度決定步驟(圖3之步驟S340)能以第一重要度獲得公式表示,第一重要度獲得方程式如下所示:In some embodiments (eg, Table 2), the importance determination step (step S340 in FIG. 3 ) can be represented by a first importance obtaining formula, and the first importance obtaining equation is as follows:
其中,I是重要度,O是原始值,是代表在與之中取最小值,B1 、B2 是無效邊界值,R為範圍值。where I is the importance, O is the original value, is represented in and Take the minimum value among them, B 1 and B 2 are invalid boundary values, and R is the range value.
在一些實施例(例如表3),重要度決定步驟(圖3之步驟S340)能以第二重要度獲得公式表示,第二重要度獲得方程式如下所示:In some embodiments (eg, Table 3), the importance determination step (step S340 in FIG. 3 ) can be represented by a second importance obtaining formula, and the second importance obtaining equation is as follows:
其中,I是重要度,O是原始值,B是無效邊界值,R為範圍值。Among them, I is the importance, O is the original value, B is the invalid boundary value, and R is the range value.
圖4為根據本案一些實施例所繪示之深度學習模型訓練系統100的示意圖。請參照圖4,在一些實施例,深度學習模型訓練系統100能搭配計量表400及影像擷取裝置500。計量表400用於顯示數值,影像擷取裝置500用於從計量表400擷取原始影像,例如擷取「計量表400用於顯示數值的區域」或「計量表400的顯示螢幕」做為原始影像。深度學習模型訓練系統100再從影像擷取裝置500獲得原始影像,因此深度學習模型訓練系統100能獲得原始影像集(即,多個原始影像)。在一些實施例,影像擷取裝置500是透過無線網路連接於深度學習模型訓練系統100。計量表400例如但不限於以機械轉動數字式或電子式顯示數值,並且計量表400例如但不限於用在電力輸送網或自來水輸送網。FIG. 4 is a schematic diagram of a deep learning
在一些實施例,深度學習模型訓練系統100除了包括處理器120及儲存裝置140之外,深度學習模型訓練系統100能更包括計量表400及影像擷取裝置500(圖中未繪示深度學習模型訓練系統100能更包括計量表400及影像擷取裝置500)。In some embodiments, in addition to the
圖5為根據本案一些實施例所繪示之計量表400的部分示意圖。請參照圖5,在一些實施例,計量表400用於顯示數值的區域能顯示4個十進位數字,也就是計量表400顯示的數值介於「0000至9999」。以圖5為例,計量表400顯示的數值為「0011」。FIG. 5 is a partial schematic diagram of a
在一些實施例,深度學習模型訓練系統100及深度學習模型訓練方法使用的深度學習模型的架構例如但不限於卷積神經網路(Convolutional Neural Network,CNN)、循環神經網路(Recurrent Neural Network,RNN)、深度神經網路(Deep Neural Network,DNN)、或yolo(You Only Look Once)。換句話說,深度學習模型訓練系統100及深度學習模型訓練方法例如但不限於依據上述所列的架構獲得置信度。In some embodiments, the architecture of the deep learning model used by the deep learning
圖6為根據本案一些實施例所繪示之物件OB與圖框的示意圖。圖7為根據本案一些實施例所繪示之交並比IOU的示意圖。請參照圖6及圖7,在一些實施例,輸入至深度學習模型的影像IM具有物件OB,深度學習模型依據物件OB於影像IM中的位置,設定第一圖框A1及第二圖框A2。深度學習模型依據第一圖框A1及第二圖框A2獲得交集區域A3及聯集區域A4,其中交集區域A3為第一圖框A1與第二圖框A2之間交集的區域,聯集區域A4為第一圖框A1與第二圖框A2之間聯集的區域。深度學習模型依據影像IM、第一圖框A1及第二圖框A2獲得影像IM的置信度,深度學習模型獲得置信度的公式例如但不限於如下所列:FIG. 6 is a schematic diagram of an object OB and a frame according to some embodiments of the present application. FIG. 7 is a schematic diagram of the intersection and ratio IOU according to some embodiments of the present application. 6 and 7, in some embodiments, the image IM input to the deep learning model has an object OB, and the deep learning model sets the first frame A1 and the second frame A2 according to the position of the object OB in the image IM . The deep learning model obtains the intersection area A3 and the union area A4 according to the first frame A1 and the second frame A2, wherein the intersection area A3 is the intersection area between the first frame A1 and the second frame A2, and the union area A4 is an area of union between the first frame A1 and the second frame A2. The deep learning model obtains the confidence of the image IM according to the image IM, the first frame A1 and the second frame A2, and the formula for obtaining the confidence by the deep learning model is, for example, but not limited to the following:
C=Pr(ob)*IOU,IOU=A3/A4。C=Pr(ob)*IOU, IOU=A3/A4.
其中,C為置信度,Pr(ob)為物件OB屬於一個類別的機率(以圖6為例,類別可為「0」,因此Pr(ob)為物件OB屬於「0」的機率),交並比IOU為交集區域A3除以聯集區域A4。Among them, C is the confidence level, Pr(ob) is the probability that the object OB belongs to a category (taking Figure 6 as an example, the category can be "0", so Pr(ob) is the probability that the object OB belongs to "0"). And the ratio of IOU is the intersection area A3 divided by the union area A4.
在一些實施例,具體而言,深度學習模型將影像IM切分為多個陣列區塊(陣列區塊共有S*S個),計算每一個陣列區塊中物件OB屬於一個類別的機率,並且從每一個陣列區塊中物件OB屬於一個類別的機率之中獲得最大值,並以此最大值做為物件OB屬於一個類別的機率。第一圖框A1為參考標準區塊(ground truth box),也就是深度學習模型依據之前訓練的結果,標記物件OB屬於一個類別的標準答案的區塊。第二圖框A2為候選區塊(bounding box),也就是深度學習模型依據輸入的影像IM,標記影像IM之中做為預測物件OB是否屬於一個類別的區塊。In some embodiments, specifically, the deep learning model divides the image IM into a plurality of array blocks (there are S*S array blocks in total), calculates the probability that the object OB in each array block belongs to a class, and The maximum value is obtained from the probability that the object OB belongs to a category in each array block, and the maximum value is used as the probability that the object OB belongs to a category. The first frame A1 is the reference standard block (ground truth box), that is, the block in which the deep learning model marks the standard answer of the object OB belonging to a category according to the result of the previous training. The second frame A2 is a candidate block (bounding box), that is, the deep learning model marks the image IM as a block that predicts whether the object OB belongs to a category according to the input image IM.
綜上,本案一些實施例的深度學習模型訓練系統、深度學習模型訓練方法及非暫態電腦可讀取儲存媒體,能依據測試值調整原始影像集以獲得測試影像集,並輸入測試影像集至第一深度學習模型進行測試而獲得測試影像集的置信度,也就是測試值的置信度。並且能依據閥值分類測試值為有效組或無效組,再從無效組中的測試值獲得訓練值。以及依據訓練值調整原始影像集以獲得訓練影像集,並輸入原始影像集及訓練影像集至第一深度學習模型進行訓練而獲得第二深度學習模型。在一些實施例,深度學習模型訓練系統及方法能利用訓練影像集對第一深度學習模型加強訓練,而不需將低於閥值的置信度所對應的全部測試值都進行訓練,因此能達到精簡深度學習模型的訓練資料量的效果,並且依然能獲得高置信度的第二深度學習模型。在一些實施例,深度學習模型訓練系統及方法能適用於辨識計量表用於顯示數值的區域之影像。To sum up, the deep learning model training system, deep learning model training method and non-transitory computer-readable storage medium of some embodiments of this case can adjust the original image set according to the test value to obtain the test image set, and input the test image set to the The first deep learning model is tested to obtain the confidence of the test image set, that is, the confidence of the test value. And the test value can be classified into the valid group or the invalid group according to the threshold value, and then the training value can be obtained from the test value in the invalid group. and adjusting the original image set according to the training value to obtain a training image set, and inputting the original image set and the training image set to the first deep learning model for training to obtain a second deep learning model. In some embodiments, the deep learning model training system and method can use the training image set to strengthen the training of the first deep learning model without training all the test values corresponding to the confidence levels lower than the threshold, thus achieving The effect of reducing the amount of training data of the deep learning model, and still obtaining a second deep learning model with high confidence. In some embodiments, deep learning model training systems and methods can be adapted to identify images of areas of meters used to display values.
100:深度學習模型訓練系統 120:處理器 140:儲存裝置 400:計量表 500:影像擷取裝置 T1:第一深度學習模型 T2:第二深度學習模型 IM:影像 OB:物件 A1:第一圖框 A2:第二圖框 A3:交集區域 A4:聯集區域 S210-S260:步驟 S310-S340:步驟100: Deep Learning Model Training System 120: Processor 140: Storage Device 400: Meter 500: Image Capture Device T1: The first deep learning model T2: Second Deep Learning Model IM: Video OB:Object A1: The first frame A2: Second frame A3: Intersection area A4: Union area S210-S260: Steps S310-S340: Steps
圖1為根據本案一些實施例所繪示之深度學習模型訓練系統的示意圖。 圖2為根據本案一些實施例所繪示之深度學習模型訓練方法的流程圖。 圖3為根據本案一些實施例所繪示之重要度獲得步驟的流程圖。 圖4為根據本案一些實施例所繪示之深度學習模型訓練系統的示意圖。 圖5為根據本案一些實施例所繪示之計量表的部分示意圖。 圖6為根據本案一些實施例所繪示之物件與圖框的示意圖。 圖7為根據本案一些實施例所繪示之交並比的示意圖。FIG. 1 is a schematic diagram of a deep learning model training system according to some embodiments of the present application. FIG. 2 is a flowchart of a deep learning model training method according to some embodiments of the present application. FIG. 3 is a flow chart of the steps of obtaining the importance according to some embodiments of the present application. FIG. 4 is a schematic diagram of a deep learning model training system according to some embodiments of the present application. FIG. 5 is a partial schematic diagram of a meter according to some embodiments of the present application. FIG. 6 is a schematic diagram of objects and frames according to some embodiments of the present application. FIG. 7 is a schematic diagram of an intersection ratio according to some embodiments of the present application.
S210-S260:步驟S210-S260: Steps
Claims (15)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW109111080A TWI750622B (en) | 2020-03-31 | 2020-03-31 | Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW109111080A TWI750622B (en) | 2020-03-31 | 2020-03-31 | Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
TW202139075A TW202139075A (en) | 2021-10-16 |
TWI750622B true TWI750622B (en) | 2021-12-21 |
Family
ID=79601195
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW109111080A TWI750622B (en) | 2020-03-31 | 2020-03-31 | Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI750622B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114510715B (en) * | 2022-01-14 | 2022-10-14 | 中国科学院软件研究所 | Method and device for testing functional safety of model, storage medium and equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109446188A (en) * | 2018-10-17 | 2019-03-08 | 大国创新智能科技(东莞)有限公司 | Follow-up auditing method and robot system based on big data and deep learning |
TW201918913A (en) * | 2017-11-01 | 2019-05-16 | 香港商阿里巴巴集團服務有限公司 | Machine processing and text correction method and device, computing equipment and storage media |
CN109767467A (en) * | 2019-01-22 | 2019-05-17 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and computer readable storage medium |
CN109948727A (en) * | 2019-03-28 | 2019-06-28 | 北京周同科技有限公司 | The training and classification method of image classification model, computer equipment and storage medium |
US10423861B2 (en) * | 2017-10-16 | 2019-09-24 | Illumina, Inc. | Deep learning-based techniques for training deep convolutional neural networks |
-
2020
- 2020-03-31 TW TW109111080A patent/TWI750622B/en active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10423861B2 (en) * | 2017-10-16 | 2019-09-24 | Illumina, Inc. | Deep learning-based techniques for training deep convolutional neural networks |
TW201918913A (en) * | 2017-11-01 | 2019-05-16 | 香港商阿里巴巴集團服務有限公司 | Machine processing and text correction method and device, computing equipment and storage media |
CN109446188A (en) * | 2018-10-17 | 2019-03-08 | 大国创新智能科技(东莞)有限公司 | Follow-up auditing method and robot system based on big data and deep learning |
CN109767467A (en) * | 2019-01-22 | 2019-05-17 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and computer readable storage medium |
CN109948727A (en) * | 2019-03-28 | 2019-06-28 | 北京周同科技有限公司 | The training and classification method of image classification model, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
TW202139075A (en) | 2021-10-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021082743A1 (en) | Video classification method and apparatus, and electronic device | |
WO2020155518A1 (en) | Object detection method and device, computer device and storage medium | |
WO2020119527A1 (en) | Human action recognition method and apparatus, and terminal device and storage medium | |
WO2017096753A1 (en) | Facial key point tracking method, terminal, and nonvolatile computer readable storage medium | |
WO2022041598A1 (en) | Remote sensing image segmentation method and system, terminal, and storage medium | |
WO2019137038A1 (en) | Method for determining point of gaze, contrast adjustment method and device, virtual reality apparatus, and storage medium | |
CN110889446A (en) | Face image recognition model training and face image recognition method and device | |
CN112183501B (en) | Depth counterfeit image detection method and device | |
WO2022237153A1 (en) | Target detection method and model training method therefor, related apparatus, medium, and program product | |
CN115294409A (en) | Video compression method, system and medium for security monitoring | |
WO2019091402A1 (en) | Method and device for age estimation | |
WO2021164550A1 (en) | Image classification method and apparatus | |
CN115661943B (en) | Fall detection method based on lightweight attitude assessment network | |
TW201250608A (en) | Image comparison system and method | |
TWI803243B (en) | Method for expanding images, computer device and storage medium | |
CN112884721B (en) | Abnormality detection method, abnormality detection system and computer-readable storage medium | |
CN107590460A (en) | Face classification method, apparatus and intelligent terminal | |
WO2020147408A1 (en) | Facial recognition model evaluation method and apparatus, and storage medium and computer device | |
WO2019120025A1 (en) | Photograph adjustment method and apparatus, storage medium and electronic device | |
TWI750622B (en) | Deep learning model training system, deep learning model training method, and non-transitory computer readable storage medium | |
CN111191685A (en) | Method for dynamically weighting loss function | |
JP2023510945A (en) | Scene identification method and apparatus, intelligent device, storage medium and computer program | |
CN108985331A (en) | Target counting method based on countermeasure generation network | |
US20230401670A1 (en) | Multi-scale autoencoder generation method, electronic device and readable storage medium | |
WO2023102844A1 (en) | Method and apparatus for determining pruning module, and computer-readable storage medium |