blocks|key|288081|text|您只需预先创建一个大小正确的数组并填充它：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|288082|frames+=+np.empty((480,+640,+3,+100))

for+k+in+xrange(nframes):
++++frames[:,:,:,k]+=+cv2.imread('frame_{}.jpg'.format(k))|code-block|syntax|javascript|288083|如果帧是以某种特定方式命名的单个jpg文件(在本例中为frame_0.jpg、frame_1.jpg等)。|288084|需要注意的是，您可以考虑使用(nframes,+480,640,3)形状的数组。|offset|length|style|CODE|288085|entityMap^0|0|0|0|E|K|0^^$0|@$1|2|3|4|5|6|7|Q|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|R|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|S|8|@]|9|@]|A|$]]|$1|I|3|J|5|6|7|T|8|@$K|U|L|V|M|N]]|9|@]|A|$]]|$1|O|3|-4|5|6|7|W|8|@]|9|@]|A|$]]]|P|$]]

You could just create an array of the correct size up-front and fill it:

<pre><code>frames = np.empty((480, 640, 3, 100))

for k in xrange(nframes):
 frames[:,:,:,k] = cv2.imread('frame_{}.jpg'.format(k))
</code></pre>

if the frames were individual jpg file that were named in some particular way (in the example, frame_0.jpg, frame_1.jpg, etc).

Just a note, you might consider using a <code>(nframes, 480,640,3)</code> shaped array, instead.

blocks|key|288115|text|您可以使用np.concatenate()指定要追加的axis，使用np.newaxis|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|288116|import+numpy+as+np
movie+=+np.concatenate((img1[:,np.newaxis],+img2[:,np.newaxis]),+axis=3)|code-block|syntax|javascript|288117|如果您正在读取多个文件：|288118|import+glob
movie+=+np.concatenate([cv2.imread(p)[:,np.newaxis]+for+p+in+glob.glob('*.jpg')],+axis=3)|288119|entityMap^0|5|G|R|4|Y|A|0|0|0|0^^$0|@$1|2|3|4|5|6|7|Q|8|@$9|R|A|S|B|C]|$9|T|A|U|B|C]|$9|V|A|W|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|X|8|@]|D|@]|E|$I|J]]|$1|K|3|L|5|6|7|Y|8|@]|D|@]|E|$]]|$1|M|3|N|5|H|7|Z|8|@]|D|@]|E|$I|J]]|$1|O|3|-4|5|6|7|10|8|@]|D|@]|E|$]]]|P|$]]

You can use <code>np.concatenate()</code> specifying which <code>axis</code> to append, using <code>np.newaxis</code>:

<pre><code>import numpy as np
movie = np.concatenate((img1[:,np.newaxis], img2[:,np.newaxis]), axis=3)
</code></pre>

If you are reading from many files:

<pre><code>import glob
movie = np.concatenate([cv2.imread(p)[:,np.newaxis] for p in glob.glob('*.jpg')], axis=3)
</code></pre>

blocks|key|70756|text|您正在询问如何向NumPy数组添加维度，以便该维度可以增长以容纳新数据。可以按如下方式添加维度：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|70757|image+=+image[...,+np.newaxis]|code-block|syntax|javascript|70758|entityMap^0|0|0^^$0|@$1|2|3|4|5|6|7|I|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|J|8|@]|9|@]|A|$E|F]]|$1|G|3|-4|5|6|7|K|8|@]|9|@]|A|$]]]|H|$]]

You're asking how to add a dimension to a NumPy array, so that that dimension can then be grown to accommodate new data. A dimension can be added as follows:

<pre><code>image = image[..., np.newaxis]
</code></pre>

blocks|key|70777|text|我遵循了这种方法：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|70778|import+numpy+as+np
import+cv2

ls+=+[]

for+image+in+image_paths:
++++ls.append(cv2.imread('test.jpg'))

img_np+=+np.array(ls)+#+shape+(100,+480,+640,+3)
img_np+=+np.rollaxis(img_np,+0,+4)+#+shape+(480,+640,+3,+100).|code-block|syntax|javascript|70779|entityMap^0|0|0^^$0|@$1|2|3|4|5|6|7|I|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|J|8|@]|9|@]|A|$E|F]]|$1|G|3|-4|5|6|7|K|8|@]|9|@]|A|$]]]|H|$]]

I followed this approach:

<pre><code>import numpy as np
import cv2

ls = []

for image in image_paths:
 ls.append(cv2.imread('test.jpg'))

img_np = np.array(ls) # shape (100, 480, 640, 3)
img_np = np.rollaxis(img_np, 0, 4) # shape (480, 640, 3, 100).
</code></pre>

blocks|key|288191|text|可替换为|type|unstyled|depth|inlineStyleRanges|entityRanges|data|288192|image+=+image[...,+np.newaxis]|code-block|syntax|javascript|288193|在@dbliss'+answer中，您还可以像这样使用numpy.expand_dims|offset|length|style|CODE|288194|image+=+np.expand_dims(image,+<your+desired+dimension>)|288195|例如(取自上面的链接)：|288196|x+=+np.array([1,+2])

print(x.shape)++#+prints+(2,)|288197|然后|288198|y+=+np.expand_dims(x,+axis=0)|288199|收益率|288200|array([[1,+2]])|288201|和|288202|y.shape|288203|给出|288204|(1,+2)|288205|entityMap|0|LINK|mutability|MUTABLE|url|https://stackoverflow.com/a/25755697/1534017|1|https://docs.scipy.org/doc/numpy/reference/generated/numpy.expand_dims.html^0|0|0|R|H|1|F|0|R|H|1|0|0|0|0|0|0|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|1I|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|1J|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|1K|8|@$I|1L|J|1M|K|L]]|9|@$I|1N|J|1O|1|1P]|$I|1Q|J|1R|1|1S]]|A|$]]|$1|M|3|N|5|D|7|1T|8|@]|9|@]|A|$E|F]]|$1|O|3|P|5|6|7|1U|8|@]|9|@]|A|$]]|$1|Q|3|R|5|D|7|1V|8|@]|9|@]|A|$E|F]]|$1|S|3|T|5|6|7|1W|8|@]|9|@]|A|$]]|$1|U|3|V|5|D|7|1X|8|@]|9|@]|A|$E|F]]|$1|W|3|X|5|6|7|1Y|8|@]|9|@]|A|$]]|$1|Y|3|Z|5|D|7|1Z|8|@]|9|@]|A|$E|F]]|$1|10|3|11|5|6|7|20|8|@]|9|@]|A|$]]|$1|12|3|13|5|D|7|21|8|@]|9|@]|A|$E|F]]|$1|14|3|15|5|6|7|22|8|@]|9|@]|A|$]]|$1|16|3|17|5|D|7|23|8|@]|9|@]|A|$E|F]]|$1|18|3|-4|5|6|7|24|8|@]|9|@]|A|$]]]|19|$1A|$5|1B|1C|1D|A|$1E|1F]]|1G|$5|1B|1C|1D|A|$1E|1H]]]]

Alternatively to 

<pre><code>image = image[..., np.newaxis]
</code></pre>

in <a href="https://stackoverflow.com/a/25755697/1534017">@dbliss' answer</a>, you can also use <a href="https://docs.scipy.org/doc/numpy/reference/generated/numpy.expand_dims.html" rel="noreferrer"><code>numpy.expand_dims</code></a> like

<pre><code>image = np.expand_dims(image, &lt;your desired dimension&gt;)
</code></pre>

For example (taken from the link above):

<pre><code>x = np.array([1, 2])

print(x.shape) # prints (2,)
</code></pre>

Then 

<pre><code>y = np.expand_dims(x, axis=0)
</code></pre>

yields

<pre><code>array([[1, 2]])
</code></pre>

and

<pre><code>y.shape
</code></pre>

gives

<pre><code>(1, 2)
</code></pre>

blocks|key|70857|text|考虑使用重塑方法的方法1和使用np.newaxis方法的方法2，它们产生相同的结果：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|70858|#Lets+suppose,+we+have:
x+=+[1,2,3,4,5,6,7,8,9]
print('I.+x',x)

xNpArr+=+np.array(x)
print('II.+xNpArr',xNpArr)
print('III.+xNpArr',+xNpArr.shape)

xNpArr_3x3+=+xNpArr.reshape((3,3))
print('IV.+xNpArr_3x3.shape',+xNpArr_3x3.shape)
print('V.+xNpArr_3x3',+xNpArr_3x3)

#Approach+1+with+reshape+method
xNpArrRs_1x3x3x1+=+xNpArr_3x3.reshape((1,3,3,1))
print('VI.+xNpArrRs_1x3x3x1.shape',+xNpArrRs_1x3x3x1.shape)
print('VII.+xNpArrRs_1x3x3x1',+xNpArrRs_1x3x3x1)

#Approach+2+with+np.newaxis+method
xNpArrNa_1x3x3x1+=+xNpArr_3x3[np.newaxis,+...,+np.newaxis]
print('VIII.+xNpArrNa_1x3x3x1.shape',+xNpArrNa_1x3x3x1.shape)
print('IX.+xNpArrNa_1x3x3x1',+xNpArrNa_1x3x3x1)|code-block|syntax|javascript|70859|我们的结果是：|70860|I.+x+[1,+2,+3,+4,+5,+6,+7,+8,+9]

II.+xNpArr+[1+2+3+4+5+6+7+8+9]

III.+xNpArr+(9,)

IV.+xNpArr_3x3.shape+(3,+3)

V.+xNpArr_3x3+[[1+2+3]
+[4+5+6]
+[7+8+9]]

VI.+xNpArrRs_1x3x3x1.shape+(1,+3,+3,+1)

VII.+xNpArrRs_1x3x3x1+[[[[1]
+++[2]
+++[3]]

++[[4]
+++[5]
+++[6]]

++[[7]
+++[8]
+++[9]]]]

VIII.+xNpArrNa_1x3x3x1.shape+(1,+3,+3,+1)

IX.+xNpArrNa_1x3x3x1+[[[[1]
+++[2]
+++[3]]

++[[4]
+++[5]
+++[6]]

++[[7]
+++[8]
+++[9]]]]|70861|entityMap^0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|M|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|N|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|O|8|@]|9|@]|A|$]]|$1|I|3|J|5|D|7|P|8|@]|9|@]|A|$E|F]]|$1|K|3|-4|5|6|7|Q|8|@]|9|@]|A|$]]]|L|$]]

Consider Approach 1 with reshape method and Approach 2 with np.newaxis method that produce the same outcome:

<pre><code>#Lets suppose, we have:
x = [1,2,3,4,5,6,7,8,9]
print('I. x',x)

xNpArr = np.array(x)
print('II. xNpArr',xNpArr)
print('III. xNpArr', xNpArr.shape)

xNpArr_3x3 = xNpArr.reshape((3,3))
print('IV. xNpArr_3x3.shape', xNpArr_3x3.shape)
print('V. xNpArr_3x3', xNpArr_3x3)

#Approach 1 with reshape method
xNpArrRs_1x3x3x1 = xNpArr_3x3.reshape((1,3,3,1))
print('VI. xNpArrRs_1x3x3x1.shape', xNpArrRs_1x3x3x1.shape)
print('VII. xNpArrRs_1x3x3x1', xNpArrRs_1x3x3x1)

#Approach 2 with np.newaxis method
xNpArrNa_1x3x3x1 = xNpArr_3x3[np.newaxis, ..., np.newaxis]
print('VIII. xNpArrNa_1x3x3x1.shape', xNpArrNa_1x3x3x1.shape)
print('IX. xNpArrNa_1x3x3x1', xNpArrNa_1x3x3x1)
</code></pre>

We have as outcome:

<pre><code>I. x [1, 2, 3, 4, 5, 6, 7, 8, 9]

II. xNpArr [1 2 3 4 5 6 7 8 9]

III. xNpArr (9,)

IV. xNpArr_3x3.shape (3, 3)

V. xNpArr_3x3 [[1 2 3]
 [4 5 6]
 [7 8 9]]

VI. xNpArrRs_1x3x3x1.shape (1, 3, 3, 1)

VII. xNpArrRs_1x3x3x1 [[[[1]
 [2]
 [3]]

 [[4]
 [5]
 [6]]

 [[7]
 [8]
 [9]]]]

VIII. xNpArrNa_1x3x3x1.shape (1, 3, 3, 1)

IX. xNpArrNa_1x3x3x1 [[[[1]
 [2]
 [3]]

 [[4]
 [5]
 [6]]

 [[7]
 [8]
 [9]]]]
</code></pre>

blocks|key|288246|text|Pythonic式的|type|unstyled|depth|inlineStyleRanges|entityRanges|data|288247|X+=+X[:,+:,+None]|offset|length|style|CODE|288248|这相当于|288249|X+=+X[:,+:,+numpy.newaxis]和X+=+numpy.expand_dims(X,+axis=-1)|288250|但是，当您明确询问有关堆叠图像的问题时，我建议您堆叠您可能在循环中收集的图像np.stack([X1,+X2,+X3])的list。|288251|如果您不喜欢维度的顺序，可以使用np.transpose()重新排列|288252|entityMap^0|0|0|H|0|0|0|Q|R|X|0|12|M|1P|4|0|G|E|0^^$0|@$1|2|3|4|5|6|7|R|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|S|8|@$D|T|E|U|F|G]]|9|@]|A|$]]|$1|H|3|I|5|6|7|V|8|@]|9|@]|A|$]]|$1|J|3|K|5|6|7|W|8|@$D|X|E|Y|F|G]|$D|Z|E|10|F|G]]|9|@]|A|$]]|$1|L|3|M|5|6|7|11|8|@$D|12|E|13|F|G]|$D|14|E|15|F|G]]|9|@]|A|$]]|$1|N|3|O|5|6|7|16|8|@$D|17|E|18|F|G]]|9|@]|A|$]]|$1|P|3|-4|5|6|7|19|8|@]|9|@]|A|$]]]|Q|$]]

Pythonic

<code>X = X[:, :, None]</code>

which is equivalent to

<code>X = X[:, :, numpy.newaxis]</code> and
<code>X = numpy.expand_dims(X, axis=-1)</code>

But as you are explicitly asking about stacking images,
I would recommend going for stacking the <code>list</code> of images <code>np.stack([X1, X2, X3])</code> that you may have collected in a loop.

If you do not like the order of the dimensions you can rearrange with <code>np.transpose()</code>

blocks|key|70899|text|这对我很有效：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|70900|image+=+image[...,+None]|code-block|syntax|javascript|70901|entityMap^0|0|0^^$0|@$1|2|3|4|5|6|7|I|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|J|8|@]|9|@]|A|$E|F]]|$1|G|3|-4|5|6|7|K|8|@]|9|@]|A|$]]]|H|$]]

This worked for me:
<pre><code>image = image[..., None]
</code></pre>

blocks|key|288299|text|您可以将stack与轴参数一起使用：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|288300|img.shape++#+h,w,3
imgs+=+np.stack([img1,img2,img3,img4],+axis=-1)+++#+-1+=+new+axis+is+last
imgs.shape+#++h,w,3,nimages|code-block|syntax|javascript|288301|例如:要将灰度转换为颜色：|288302|>>>+d+=+np.zeros((5,4),+dtype=int)++#+5x4
>>>+d[2,3]+=+1

>>>+d3.shape
Out[30]:+(5,+4,+3)

>>>+d3+=+np.stack([d,d,d],+axis=-2)++#+5x4x3+++-1=as+last+axis
>>>+d3[2,3]
Out[32]:+array([1,+1,+1])|288303|entityMap^0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|M|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|N|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|O|8|@]|9|@]|A|$]]|$1|I|3|J|5|D|7|P|8|@]|9|@]|A|$E|F]]|$1|K|3|-4|5|6|7|Q|8|@]|9|@]|A|$]]]|L|$]]

You can use stack with the axis parameter:
<pre><code>img.shape # h,w,3
imgs = np.stack([img1,img2,img3,img4], axis=-1) # -1 = new axis is last
imgs.shape # h,w,3,nimages
</code></pre>
For example: to convert grayscale to color:
<pre><code>&gt;&gt;&gt; d = np.zeros((5,4), dtype=int) # 5x4
&gt;&gt;&gt; d[2,3] = 1

&gt;&gt;&gt; d3.shape
Out[30]: (5, 4, 3)

&gt;&gt;&gt; d3 = np.stack([d,d,d], axis=-2) # 5x4x3 -1=as last axis
&gt;&gt;&gt; d3[2,3]
Out[32]: array([1, 1, 1])
</code></pre>

I'm starting off with a numpy array of an image.

<pre><code>In[1]:img = cv2.imread('test.jpg')
</code></pre>

The shape is what you might expect for a 640x480 RGB image.

<pre><code>In[2]:img.shape
Out[2]: (480, 640, 3)
</code></pre>

However, this image that I have is a frame of a video, which is 100 frames long. Ideally, I would like to have a single array that contains all the data from this video such that <code>img.shape</code> returns <code>(480, 640, 3, 100)</code>.

What is the best way to add the next frame -- that is, the next set of image data, another 480 x 640 x 3 array -- to my initial array?

How can I add new dimensions to a Numpy array?

我从一个numpy数组的图像开始。In[1]:img = cv2.imread('test.jpg')该形状可能是您对640x480 RGB图像的期望。In[2]:img.shapeOut[2]: (480, 640, 3)然而，我得到的这张图片是一段视频的一帧，它有100帧长。理想情况下，我希望有一个包含此视频中所有...

问如何向Numpy数组添加新的维数？
EN

社区

活动

资源

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问如何向Numpy数组添加新的维数？EN

社区

活动

资源

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问如何向Numpy数组添加新的维数？
EN