**Preface to "AI-Based Transportation Planning and Operation"**

Due to drastic urbanization, cities around the world are facing transportation-induced problems such as congestion, accidents, and air pollution. Transportation planning and operation provide opportunities to satisfy the demand for the movement of people and goods in a safe, economical, convenient, and sustainable manner. Previous studies of transportation planning and operation have depended upon econometrics or engineering-based modeling, which cannot fully incorporate the power of data-driven or artificial intelligence (AI)-based approaches. Recently, AI technologies, such as supervised/unsupervised learning, reinforcement learning, and Bayesian modeling, have boosted the capability of dealing with the complexity and high nonlinearity in the problems of transportation planning and operations. More specifically, AI-based technologies in decision-making, planning, modeling, estimation, and control have facilitated the process of transportation planning and operations. This Special Issue provides an academic platform to publish high-quality articles on the applications of innovative AI algorithms to transportation planning and operation. The topics of the articles encompass AI technologies for traffic surveillance, vehicle emission reduction, traffic safety, congestion management, traffic speed forecasting, and ride sharing strategy.

> **Keemin Sohn** *Editor*

*Article*

### **Measuring Tra**ffi**c Volumes Using an Autoencoder with No Need to Tag Images with Labels**

### **Seungbin Roh, Johyun Shin and Keemin Sohn \***

Department of Urban Engineering, Chung-Ang University, 84 Heukseok-ro, Dongjak-gu, Seoul 156-756, Korea; sbr444@cau.ac.kr (S.R.); olfy1021@cau.ac.kr (J.S.)

**\*** Correspondence: kmsohn@cau.ac.kr

Received: 31 March 2020; Accepted: 24 April 2020; Published: 25 April 2020

**Abstract:** Almost all vision technologies that are used to measure traffic volume use a two-step procedure that involves tracking and detecting. Object detection algorithms such as YOLO and Fast-RCNN have been successfully applied to detecting vehicles. The tracking of vehicles requires an additional algorithm that can trace the vehicles that appear in a previous video frame to their appearance in a subsequent frame. This two-step algorithm prevails in the field but requires substantial computation resources for training, testing, and evaluation. The present study devised a simpler algorithm based on an autoencoder that requires no labeled data for training. An autoencoder was trained on the pixel intensities of a virtual line placed on images in an unsupervised manner. The last hidden node of the former encoding portion of the autoencoder generates a scalar signal that can be used to judge whether a vehicle is passing. A cycle-consistent generative adversarial network (CycleGAN) was used to transform an original input photo of complex vehicle images and backgrounds into a simple illustration input image that enhances the performance of the autoencoder in judging the presence of a vehicle. The proposed model is much lighter and faster than a YOLO-based model, and accuracy of the proposed model is equivalent to, or better than, a YOLO-based model. In measuring traffic volumes, the proposed approach turned out to be robust in terms of both accuracy and efficiency.

**Keywords:** autoencoder; deep learning; traffic volume; vehicle counting; CycleGAN
