⚡ 本页包含 AI 生成的分析内容，仅供参考
提出一种数据压缩型QVGA图像传感器，采用1.5b/2.75b的对数梯度读出和多尺度读出架构，专为始终开启的目标检测应用设计。解决了传统HOG特征提取系统中传感器与后端检测器之间数据量大、能耗不平衡的问题，通过前端压缩降低后端处理功耗。
CA 1 2 Histograms of Oriented Gradients (HOG) are attractive features for object detection in embedded vision applications, as they provide a good trade-off between complexity and detection accuracy. A custom 8b CMOS imager that computes these features on-chip consumes only 52pJ/pixel [1]. However, a complete system also requires a backend detection algorithm, which consumes 940pJ/pixel in an optimized implementation [2]. As shown in the system study of [3], this imbalance mostly stems from the large amount of data seen by the detector. To remedy this issue, the work of [3] studies a feature-extraction approach that aggressively log-quantizes the data, thereby eliminating unnecessary illuminationrelated bits from the histograms. The custom log-gradient image sensor described in this paper demonstrates this concept in CMOS. It consumes 127pJ/pixel and
Christopher Young1, Alex Omid-Zohoor1,2, Pedram Lajevardi3, Boris Murmann1
Stanford University, Stanford, CA *now with K-Motion Interactive, San Francisco, CA 3 Robert Bosch, Sunnyvale