TY - JOUR
T1 - Convolutional Neural Network Image Classification Based on Different Color Spaces
AU - Xian, Zixiang
AU - Huang, Rubing
AU - Towey, Dave
AU - Yue, Chuan
PY - 2025/2
Y1 - 2025/2
N2 - Although Convolutional Neural Networks (CNNs) have achieved remarkable success in image classification, most CNNs use image datasets in the Red-Green-Blue (RGB) color space (one of the most commonly used color spaces). The existing literature regarding the influence of color space use on the performance of CNNs is limited. This paper explores the impact of different color spaces on image classification using CNNs. We compare the performance of five CNN models with different convolution operations and numbers of layers on four image datasets, each converted to nine color spaces. We find that color space selection can significantly affect classification accuracy, and that some classes are more sensitive to color space changes than others. Different color spaces may have different expression abilities for different image features, such as brightness, saturation, hue, etc. To leverage the complementary information from different color spaces, we propose a pseudo-Siamese network that fuses two color spaces without modifying the network architecture. Our experiments show that our proposed model can outperform the single-color-space models on most datasets. We also find that our method is simple, flexible, and compatible with any CNN and image dataset.
AB - Although Convolutional Neural Networks (CNNs) have achieved remarkable success in image classification, most CNNs use image datasets in the Red-Green-Blue (RGB) color space (one of the most commonly used color spaces). The existing literature regarding the influence of color space use on the performance of CNNs is limited. This paper explores the impact of different color spaces on image classification using CNNs. We compare the performance of five CNN models with different convolution operations and numbers of layers on four image datasets, each converted to nine color spaces. We find that color space selection can significantly affect classification accuracy, and that some classes are more sensitive to color space changes than others. Different color spaces may have different expression abilities for different image features, such as brightness, saturation, hue, etc. To leverage the complementary information from different color spaces, we propose a pseudo-Siamese network that fuses two color spaces without modifying the network architecture. Our experiments show that our proposed model can outperform the single-color-space models on most datasets. We also find that our method is simple, flexible, and compatible with any CNN and image dataset.
KW - Accuracy
KW - Brightness
KW - Convolution
KW - Convolutional Neural Network (CNN)
KW - Fuses
KW - Image color analysis
KW - Network architecture
KW - Neural networks
KW - Color space
KW - Image classification
KW - pseudo-Siamese network
UR - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=pure_ris_china&SrcAuth=WosAPI&KeyUT=WOS:001314325700012&DestLinkType=FullRecord&DestApp=WOS_CPL
U2 - 10.26599/TST.2024.9010001
DO - 10.26599/TST.2024.9010001
M3 - Article
SN - 1007-0214
VL - 30
SP - 402
EP - 417
JO - Tsinghua Science and Technology
JF - Tsinghua Science and Technology
IS - 1
ER -