1. GRAPH_PATH：graph檔案路徑；

2. IMAGE_PATH：要分類的圖片的路徑；

3. IMAGE_DIM：由選擇的神經網路定義的影象尺寸；  例：GoogLeNet uses 224x224 pixels, AlexNet uses 227x227 pixels

4. IMAGE_STDDEV：由選擇的神經網路定義的標準差（標度值） 例：GoogLeNet uses no scaling factor, InceptionV3 uses 128 (stddev = 1/128)

5. IMAGE_MEAN: 平均減法是深度學習中常用的一種技術，用於對資料進行中心處理  

例：ILSVRC dataset, the mean is B = 102 Green = 117 Red = 123

使用NCS做影象分類的5個步驟：

 從mvnc庫中引入mvncapi模組

import mvnc.mvncapi as mvnc

Step 1:

   將NCS插入應用處理器（Ubuntu膝上型電腦/桌上型電腦）USB埠時，它將自身列為USB裝置。將呼叫API來查詢列舉的NCS裝置:

    # Look for enumerated Intel Movidius NCS device(s); quit program if none found.

devices = mvnc.EnumerateDevices()

if len( devices ) == 0:

print( 'No devices found' )

quit()

  如果插入了多個NCS，還需要選擇一個NCS並開啟：

    # Get a handle to the first enumerated device and open it

device = mvnc.Device( devices[0] )

device.OpenDevice()

Step2：載入graph檔案到NCS

    # Read the graph file into a buffer

with open( GRAPH_PATH, mode='rb' ) as f:

blob = f.read()

# Load the graph buffer into the NCS

graph = device.AllocateGraph( blob )

Step3: 將單個影象載入到Intel Movidius NCS上以執行推理

影象預處理：

1. 調整影象大小/裁剪影象以匹配預先訓練的網路定義的尺寸。

        GoogLeNet uses 224x224 pixels, AlexNet uses 227x227 pixels.

2. 每個通道的平均值（藍色，綠色和紅色）從整個資料集中減去。這是深度學習中常用的一種技術，可以集中資料。

3. 將影象轉換為半精度浮點數（fp16）陣列，並使用LoadTensor函式呼叫將影象載入到NCS上。skimage庫可以在一行程式碼中完成此操作。

# Read & resize image [Image size is defined during training]

img = print_img = skimage.io.imread( IMAGES_PATH )

img = skimage.transform.resize( img, IMAGE_DIM, preserve_range=True )

# Convert RGB to BGR [skimage reads image in RGB, but Caffe uses BGR]

img = img[:, :, ::-1]

# Mean subtraction & scaling [A common technique used to center the data]

img = img.astype( numpy.float32 )

img = ( img - IMAGE_MEAN ) * IMAGE_STDDEV

# Load the image as a half-precision floating point array

graph.LoadTensor( img.astype( numpy.float16 ), 'user object' )

Step4: 從NCS讀取並列印推理結果

    # Get the results from NCS

output, userobj = graph.GetResult()

# Print the results

print('\n------- predictions --------')

labels = numpy.loadtxt( LABELS_FILE_PATH, str, delimiter = '\t' )

order = output.argsort()[::-1][:6]

for i in range( 0, 5 ):

print ('prediction ' + str(i) + ' is ' + labels[order[i]])

# Display the image on which inference was performed

skimage.io.imshow( IMAGES_PATH )

skimage.io.show( )

Step 5: 解除安裝圖形並關閉裝置

為了避免記憶體洩漏和/或分段錯誤，我們應該關閉所有開啟的檔案或資源並釋放所有使用的記憶體。

graph.DeallocateGraph()

device.CloseDevice()