| Sign In | Join Free | My insurersguide.com |
|
Storage : 8GB/16GB eMMC
Power : 5V/1A
Output Format : Camera(IR): RAW Camera(RGB): RAW
Recommended Database : 10,000
Module Size : 84.0mm × 22.45mm × 19.35mm
Processor : Quad-core ARM Cortex-A7 32-bit, 1.5GHz, with integrated NEON and FPU Each core has a 32KB I-cache and 32KB D-cache, plus 512KB shared L2 cache Based on RISC-V MCU
Interface : Camera(IR): MIPI Camera(RGB): MIPI
Maximum Database : 100,000
Recommended Face Recognition Angles : Yaw: ≤ ±30° Pitch: ≤ ±30° Roll: ≤ ±30°
Face Comparison : Feature Extraction Time: ~25 ms Single Comparison Time: ~0.0115 ms
Video decoding : 4KH.264/H.26530fps 3840x2160@30encoding+3840x2160@30fpsdecoding
Image Sensors : Camera(IR): GC2053 Camera(RGB): GC2093
Pixel Size : Camera(IR): 2.8 μm Camera(RGB): 2.8 μm
Recommended Image : 720P
Video encoding : 4KH.264/H.26530fps 3840x2160@30fps+720p@30fpsencoding
Sensor Size : Camera(IR): 1 / 2.9 Camera(RGB): 1 / 2.9
System support : Linux
Operating humidity : 10%~90%
Resolution : Camera(IR): Center 800 Edge 600 Camera(RGB): Center 800 Edge 600
Face Recognition Accuracy : Standard Testing Environment, 10,000-person Database: Without Mask: False Acceptance Rate: 0.01%; Recognition Accuracy: 99% With Mask: False Acceptance Rate: 0.01%; Recognition Accuracy: 95%
Enclosure Design : Aluminum alloy material with serrated heat sink back cover for efficient cooling
Lens : Camera(IR): 4P Camera(RGB): 4P
Liveness Detection : Monocular Liveness Detection Time: ~45 ms Binocular Liveness Detection Time: ~15 ms
NPU : Up to 2.0 Tops performance, supports INT8/INT16, strong network model compatibility, RKNN model conversion tool available for converting common AI framework models (e.g., Caffe, Darknet, MXNet, ONNX, PyTorch, TensorFlow, TFLite) and algorithm support
Face Detection : Face Detection Time: ~23 ms Face Tracking Time: ~7 ms
Memory : 1GB/2GBDDR4
Filter Wavelength : Camera(IR): 850 nm Camera(RGB): 650 nm
Payment Terms : T/T
Optical Distortion : Camera(IR): ≤0.5% Camera(RGB): ≤0.5%
Focal Length : Camera(IR): F2.0/4.3mm Camera(RGB): F2.0/4.3mm
Host computer chip : RV1126
Focusing Distance : Camera(IR): 80 cm Camera(RGB): 80 cm
Model Number : JP1126
Place of Origin : China
MOQ : Negotiable
Price : Negotiable
Supply Ability : 200+/day
Delivery Time : 5-8 work days
Operating temperature : -10℃~60℃
Power Consumption : Typical Power Consumption: 2.8W (5V, 560mA) Maximum Power Consumption: 4.3W (5V, 860mA) Minimum Power Consumption: 0.71W (5V, 142mA) Power Supply Recommendation: 5V/1.2A or higher
Field of View : Camera(IR): D70°H62°V38° Camera(RGB): D70°H62°V38°
Minimum Face Size for Recognition : Without Liveness Detection: 50 x 50 pixels With Liveness Detection: 90 x 90 pixels)
JP1126 Intelligent Dual-Lens Camera Module Up to 2.0 Tops performance, supports INT8/INT16 5V/1A
JP1126 Intelligent Dual-Lens Camera Module Features:
JP1126 Intelligent Dual-Lens Camera Module Parameter:
|
Processor:
|
Quad-core ARM Cortex-A7 32-bit, 1.5GHz, with integrated NEON and FPU
Each core has a 32KB I-cache and 32KB D-cache, plus 512KB shared L2 cache
Based on RISC-V MCU
|
|
NPU:
|
Up to 2.0 Tops performance, supports INT8/INT16, strong network model compatibility,
RKNN model conversion tool available for converting common AI framework models (e.g.,
Caffe, Darknet, MXNet, ONNX, PyTorch, TensorFlow, TFLite) and algorithm support
|
|
Memory:
|
1GB/2GBDDR4
|
|
Storage:
|
8GB/16GB eMMC
|
|
Video encoding:
|
4KH.264/H.26530fps
3840x2160@30fps+720p@30fpsencoding
|
|
Video Decoding:
|
4KH.264/H.26530fps
3840x2160@30encoding+3840x2160@30fpsdecoding
|
|
System support:
|
Linux
|
|
Power:
|
5V/1A
|
|
Image Sensors:
|
GC2053
GC2093
|
| Module Board Dimensions: | 80* 16* 17.6mm (L* W* H) |
|
Resolution:
|
1920*1080
|
|
Pixel Size:
|
2.8 μm
|
|
Interface:
|
MIPI
|
|
Focal Length:
|
F2.0/4.3mm
|
| Maximum Database: |
100,000
|
|
Face Recognition
Accuracy:
|
Standard Testing Environment, 10,000-person Database:
Without Mask:
False Acceptance Rate: 0.01%; Recognition Accuracy: 99%
With Mask:
False Acceptance Rate: 0.01%; Recognition Accuracy: 95%
|



|
|
Binocular Face Recognition 3D Stereo Vision Camera Module Images |