4D Radar Imaging and Camera Fusion for Road Crossing Detection and Classification Using Deep Learning
Bibliographic record
Abstract
This paper presents the development of an object detection and classification system for road crossing areas, integrating 4D radar imaging and a mono-camera dataset with a deep-learning neural network. The system utilizes deep neural networks implemented via Keras and TensorFlow to detect and classify multiple targets, including pedestrians, cars, buses, and trucks. At the core of this work is Retina-4F, a multi-chip radar imaging system developed by Smart Radar System, which offers high-resolution object detection and localization capabilities. Retina-4F provides real-time 4D information on detected objects, operating in a cascading architecture with three transmitters and four receivers per chip. Two road-crossing scenes were simulated to collect data, generating a point cloud dataset labeled with target classes for neural network training and testing. Data from two main sensors—Retina-4F and a mono-camera—were pre-processed using DBSCAN and YOLOv7 for enhanced accuracy. Operating at 77 GHz, Retina-4F was tested in two road environments, generating a dataset with approximately 10,000 frames. The deep learning model demonstrated an accuracy of 84% in classifying multiple targets, including cars, pedestrians, buses, and trucks. The fusion of radar point cloud data with visual sensor data proved effective, showing strong results in distinguishing target types. ABSTRAK: Kertas ini membentangkan pembangunan sistem pengesanan dan pengelasan objek untuk kawasan lintasan jalan raya, menggabungkan pengimejan radar 4D dan set data mono-kamera dengan rangkaian neural pembelajaran mendalam. Sistem ini menggunakan rangkaian neural mendalam yang dilaksanakan melalui Keras dan TensorFlow untuk mengesan dan mengelaskan pelbagai sasaran, termasuk pejalan kaki, kereta, bas, dan trak. Inti daripada kajian ini adalah Retina-4F, sistem pengimejan radar berbilang cip yang dibangunkan oleh Smart Radar System, yang menawarkan keupayaan pengesanan objek dan penentuan lokasi resolusi tinggi. Retina-4F menyediakan maklumat 4D masa nyata mengenai objek yang dikesan, beroperasi dengan tiga pemancar dan empat penerima bagi setiap cip dalam seni bina kaskad. Dua adegan lintasan jalan disimulasikan untuk mengumpul data, menghasilkan set data awan titik yang dilabel dengan kelas sasaran untuk latihan dan ujian rangkaian neural. Data daripada dua sensor utama—Retina-4F dan mono-kamera—dipra-proses menggunakan DBSCAN dan YOLOv7 untuk meningkatkan ketepatan. Beroperasi pada 77 GHz, Retina-4F diuji dalam dua persekitaran jalan yang berbeza, menghasilkan set data dengan kira-kira 10,000 bingkai. Model pembelajaran mendalam menunjukkan ketepatan sebanyak 84% dalam mengelaskan pelbagai sasaran, termasuk kereta, pejalan kaki, bas, dan trak. Penggabungan data awan titik radar dengan data sensor visual terbukti berkesan, menunjukkan hasil yang kuat dalam membezakan antara jenis sasaran.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".