Scene text detection using enhanced extremal region and convolutional neural network

F Naiemi, V Ghods, H Khalesi - Multimedia Tools and Applications, 2020 - Springer
… a method to detect texts in natural images. The proposed … in the preprocessing stage: image
mapping to distinct R, G and … and Gaussian low pass filter to remove the impulse noise and …

A deep learning approach for robust, multi-oriented, and curved text detection

R Ranjbarzadeh, S Jafarzadeh Ghoushchi, S Anari… - Cognitive …, 2022 - Springer
… from a selected region inside the text images for recognizing … region of occurring, and text
region formation are eliminated [50]. … Grad-CAM method produces an attention map by utilizing …

[HTML][HTML] Graphical image region extraction with k-means clustering and watershed

S Jardim, J António, C Mora - Journal of Imaging, 2022 - mdpi.com
… resulting infographical map for extracting the proposed regions, … i the number of iterations
needed to achieve convergence. … removing unnecessary visual elements from images, such as …

[HTML][HTML] Semantic segmentation deep learning for extracting surface mine extents from historic topographic maps

AE Maxwell, MS Bester, LA Guillen, CA Ramezan… - Remote Sensing, 2020 - mdpi.com
… Further, when reducing the number of topographic maps used to derive training image
same training epoch, even after many training iterations. Also, the model generalized poorly to …

Enriching the metadata of map images: a deep learning approach with GIS-based data augmentation

Y Hu, Z Gui, J Wang, M Li - International Journal of Geographical …, 2022 - Taylor & Francis
… is a map image, which can be scanned from a paper map or … to remove noise images with
the goal of keeping 50 images for … text annotations, which make it difficult to generate map

[HTML][HTML] Individual tree crown segmentation and crown width extraction from a heightmap derived from aerial laser scanning data using a deep learning framework

C Sun, C Huang, H Zhang, B Chen, F An… - Frontiers in plant …, 2022 - frontiersin.org
… aim of accurately delineating ITC boundaries with overlapping … , image-to-image translation
is utilized to learn the mapping … In each iteration step, the fake images yielded by generator G …

[HTML][HTML] A robust and fast method for sidescan sonar image segmentation based on region growing

X Wang, L Wang, G Li, X Xie - Sensors, 2021 - mdpi.com
… SNR of a side scan sonar image, remove the speckle noise, … -time iterative growth of the
seabed reverberation area. The … ground-truth segmentation maps of the real sonar images are …

[HTML][HTML] An enhanced binarization framework for degraded historical document images

W Xiong, L Zhou, L Yue, L Li, S Wang - EURASIP Journal on Image and …, 2021 - Springer
… to preserve text stroke connectivity and eliminate isolated noise. … classifier is then applied to
iteratively classify the uncertain pixels … parameter settings produces a cleaner text edge map. …

Combining deep learning and mathematical morphology for historical map segmentation

Y Chen, E Carlinet, J Chazalon, C Mallet… - … Conference on Discrete …, 2021 - Springer
… focus on boundary detection, ie, a middle-level image task that … After several iterations, the
probability of text pixels will vanish, … [29] also allow to eliminate some minima based on their …

Historical document image binarization: A review

C Tensmeyer, T Martinez - SN Computer Science, 2020 - Springer
… document images, it is typical to map foreground text pixels … The winner of DIBCO 2009
proposed iterative polynomial … output binary image to remove noise. Generally, this involves …