Stack Overflow
Stack Overflow

Questions tagged [data-augmentation]

Data augmentation

Data augmentation is the technique of increasing the size of data used for training a model. It also helps in preventing overfitting.

Some of the most common data augmentation techniques for images include:

  • Scaling
  • Cropping
  • Flipping
  • Rotation
  • Color jittering
465 questions
0
votes
0 answers

Single image augmentation using different magnitudes of noise

Is it practical to create a dataset with noise off of just one image? Recently, I have asked here with regards to adding noise to images. I am aware that convolutional neural networks require datasets with thousands of images. However, my goal is to…
Jerome Ariola
  • 135
  • 1
  • 11
0
votes
1 answer

Will the augmented data set really improve the ML model

We previously had 411 size data set. With that we got 70 % of the objects detected. We augmented it to 5000 by implementing rotations and different levels of brightnesses. But the accuracy dropped to 40%. We don't know where it went wrong.
Phaniraj Kattamuru
  • 1
  • 2
0
votes
1 answer

trouble creating and saving augmented images using imgaug

Using python v 3.7.3, pytorch v 0.4.1, imgaug 0.3.0, windows 10, Jupyter Notebook I am trying to iterate through several folders containing images, augment each image 6 times, then save a hard copy of each augmented image inside that folder. I am…
Sanglang
  • 81
  • 1
  • 10
0
votes
0 answers

imgaug: is it possible to keep the augmenter at the current transformation?

I'm working at augmenting a dataset made of sequences of frames: i'd like to apply a particular transformation to all the frames in a sequence, while normally imgaug applies a random transformation (within the given parameters) for each image fed…
erolo4454
  • 1
  • 1
0
votes
0 answers

getting error while performing data augmentation whien i try to import image_to_array in keras

ImportError: cannot import name 'image_to_array' from 'keras.preprocessing.image' (C:\Users\shkatta\New folder\lib\site-packages\keras\preprocessing\image.py) i am not able to import the above error when i am importing image_to array in…
sheshank
  • 29
  • 4
0
votes
2 answers

Increasing dataset size using imgaug

I am merging two different datasets containing images into one dataset. One of the datasets contains 600 images in the training set. The other dataset contains only 90-100 images. I want to increase the size of the latter dataset by using the…
Sanglang
  • 81
  • 1
  • 10
0
votes
1 answer

Image Data Generator for a folder of images

I have 7 labeled classes all with varying quantities of images in them (ranging from 2000-20000). I know in keras when using the model.fit I can change how many times each folder of labeled images is read in. Instead, I would like to compare the…
cdr
  • 21
  • 1
  • 7
0
votes
2 answers

Cast ImageDataGenerator Data Output

I'm writing a network for Image Segmentation. I have my ImageDataGenerator for my masks (which are RGB images with only 0 and 255 as values, black and white) which is: train_mask_data_gen = ImageDataGenerator(rotation_range=10, …
Mattia Surricchio
  • 1,362
  • 2
  • 21
  • 49
0
votes
0 answers

Changing the elements of numpy.ndarray of Keras ImageDataGenerator

I am using Keras ImageDataGenerator for data augmentation. After generating ImDatagenerator as follows, datagen = ImageDataGenerator( rescale=1./255.) DG = datagen.flow_from_directory( directory='Path_To_Dataset', …
Hadi Zand
  • 1
  • 1
0
votes
1 answer

Error when checking target: expected dense_34 to have 2 dimensions, but got array with shape (64, 10, 2)

I see that similar questions have been answered and this has helped me to realize that the input is not what the model expects, but nowhere have I been able to find how to correct this. My question is why is it expecting 2 dimensions and what can…
Jacob Myer
  • 479
  • 5
  • 22
0
votes
1 answer

Data augmentation function not correct

Just a head's up: I'm new to this, so please go gentle. I'm trying to create a function that will shift every image in the MNIST dataset and add the shifted image to the original dataset, effectively doubling the dataset size. My code (a warning, it…
Yehuda
  • 1,787
  • 2
  • 15
  • 49
0
votes
1 answer

Method of visualizing fit of cross validated model

How do I go about writing code to visualize the progress of my accuracy and loss development over training when using cross validation? Normally I would assign the variable name 'history' to the fit function when training the model, but in the case…
Janne
  • 47
  • 5
0
votes
1 answer

Is it okay if we augment the data first then randomly choose the data and split the data afterward?

I am doing a science project about classifying medical images but I do not have a lot of data so, is it okay if I augment the data first then randomly select the data to keep and split the kept data afterward? At first, my teacher told me to augment…
Waranthorn Chansawang
  • 89
  • 1
  • 6
0
votes
1 answer

Is it feasible to do random cropping as a data augmentation technique in the context of multi-label image classification?

I have read 2 top-ranking solutions in kaggle concerning multi-label image classification. In both of the competitions I read, random cropping was performed. To me, this seems like a bad move to make because we could have a mismatch between the…
user
  • 93
  • 10
0
votes
1 answer

What are some good data augmentation techniques for document images?

I have 1000 resume in png format and I am implementing MaskRcnn for object detection. What data augmentation techniques can I use to improve the mask Rcnn performance?
hR 312
  • 824
  • 1
  • 9
  • 22
1 2 3
…
30 31