Save image only from figure as png in a loop on matplotlib with same dimensions

Question

I am working in computer vision, and I am trying to convert Microsoft Coco dataset into png images so that I can use them directly on Caffe.

I modified some function of their API so that I have the correct segmentation colors (each class id has (id,id,id) color in rgb, and background is (0,0,0)). Here is the modified function:

def showAnns(self, anns):
    """
    Display the specified annotations.
    :param anns (array of object): annotations to display
    :return: None
    """
    if len(anns) == 0:
        return 0
    if 'segmentation' in anns[0]:
        datasetType = 'instances'
    elif 'caption' in anns[0]:
        datasetType = 'captions'
    if datasetType == 'instances':
        ax = plt.gca()
        polygons = []
        color = []
        # sort annotations from biggest to smallest to avoid occlusions
        anns.sort(key=lambda x: x['area'], reverse=True)

        for ann in anns:

            pixelvalue = ann['category_id']/255.0
            c = [pixelvalue, pixelvalue, pixelvalue]
            if type(ann['segmentation']) == list:
                # polygon
                for seg in ann['segmentation']:
                    poly = np.array(seg).reshape((len(seg)/2, 2))
                    polygons.append(Polygon(poly, True,alpha=1))
                    color.append(c)
            else:
                # mask
                t = self.imgs[ann['image_id']]
                if type(ann['segmentation']['counts']) == list:
                    rle = mask.frPyObjects([ann['segmentation']], t['height'], t['width'])
                else:
                    rle = [ann['segmentation']]
                m = mask.decode(rle)
                img = np.ones( (m.shape[0], m.shape[1], 3) )
                color_mask = c
                for i in range(3):
                    img[:,:,i] = color_mask[i]
                ax.imshow(np.dstack( (img, m*0.5) ))

        p = PatchCollection(polygons, facecolors=color, edgecolors=(1,1,1,1), linewidths=1, alpha=1)
        ax.add_collection(p)
    elif datasetType == 'captions':
        for ann in anns:
            print ann['caption']

This code is called in my python script below:

from pycocotools.coco import COCO
import numpy as np
import skimage.io as io
import matplotlib.pyplot as plt

dataDir='..'
dataType='train2014'
annFile='%s/annotations/instance`enter code here`s_%s.json'%(dataDir,dataType)

# initialize COCO api for instance annotations
coco=COCO(annFile)

#display COCO categories and supercategories
cats = coco.loadCats(coco.getCatIds())

# get all images
catIds = coco.getCatIds();
imgIds = coco.getImgIds()
imgs = coco.loadImgs(imgIds)
imgs.sort(key=lambda x: x['id'])

for img in imgs:

    # load and display image
    I = io.imread('%s/images/%s/%s'%(dataDir,dataType,img['file_name']))
    # 0 all values
    I[:] = 0

    axes = plt.gca()
    axes.set_frame_on(False)
    axes.set_xticks([]); axes.set_yticks([])

    plt.axis('off')
    plt.imshow(I)

    # load and display instance annotations
    annIds = coco.getAnnIds(imgIds=img['id'], catIds=catIds, iscrowd=None)
    anns = coco.loadAnns(annIds)
    coco.showAnns(anns)
    path='vocstyle_cocoimages/'+str(img['file_name'])[:-4]+'.png'
    plt.savefig(path, bbox_inches='tight', pad_inches=0)
    plt.clf()

So here is my problem here: When I do this, I obtain a save png with a different size from the original one. I would like to have exactly the same dimensions, since it is important for evaluation: any misplaced pixel will lower the accuracy in the deep learning phase.

I haven't found a proper way to save the "image" part of the figure stripped of any other thing in the figure. Let's say I have a 640*480 image input, I would like to have the same size as an ouput.

Thank you for reading

Answer 1

Use a counter, and append that to your path string:

COUNT = 0

for img in imgs:

    # load and display image
    I = io.imread('%s/images/%s/%s'%(dataDir,dataType,img['file_name']))
    # 0 all values for black background
    I[:] = 0
    plt.imshow(I)
    # load and display instance annotations
    annIds = coco.getAnnIds(imgIds=img['id'], catIds=catIds, iscrowd=None)
    anns = coco.loadAnns(annIds)
    coco.showAnns(anns)
    print img['file_name']
    #path='vocstyle_cocoimages/'+str(img['file_name'])[:-4]+'.png'
    #your new line
    path='vocstyle_cocoimages/'+str(img['file_name'])[:-4]+ '_' + str(COUNT) + '.png'
    plt.savefig(path, bbox_inches='tight', pad_inches=0)
    #increment counter
    COUNT = COUNT + 1