Convolutional Neural Network Based Vehicle Classification in Adverse Illuminous Conditions for Intelligent Transportation Systems

Table 2

Classification accuracy of the proposed system with different dimensional feature vectors on self-constructed dataset.

No. of FCs	fc1 out-features	fc2 in-features	fc2 out-features	Accuracy
1	6	—	—	95.14%
2	1536	1536	6	99.74%
2	1024	1024	6	99.68%
2	768	768	6	99.51%