Nowadays AI seems become one of main trend in IP video surveillance technology, Huawei recently launched 2-megapixel face recognition security camera model - M1221-F. The product adopts Hisilicon processor which featuring Extra265 codec, T-Shot face capture and Mate10 noise reduction algorithm have greatly improved image quality; it also has received more than 108 security certificates, iHealth image smart diagnosis, IPCA network quality analysis, eSafe security system to improve efficiency of surveillance. The camera adopts durable aluminum housing design, special heat dissipation design helps to cool system can provide higher reliability, including multi-grade protection circuit design to ensure reliable performance even under severe weather conditions.
Huawei M1221-F is an intelligent camera for face image checkpoints, helping to build a face image system with accurate recognition, fast snapshot, high efficiency, and stability. Face detection and snapshot is mainly based on computer vision, integrating multiple technologies, such as image processing, pattern recognition, and deep learning. It can analyze massive numbers of videos in real time, filter out useless data, and extract valuable information. The dynamic face image algorithm designed for densely populated scenes is used to detect, track, and optimize dynamic face images for the optimal snapshots. It accurately captures dynamic face images in densely populated scenes, entrances and exits, facilitating the transformation of public security, commercial, and judicial industries towards intelligent systems.
Huawei M1221-F Smart IPC leverages high performance 2-megapixel 1/1.8” size progressive scan CMOS image sensor, it adopts Hisilicon (subsidiary of Huawei) SoC with embedded Extra265 algorithm, ultra high efficient codec can save up to 80% bandwidth comparing with traditional H.264/AVC standard. Additionally, it employs smart I-frame technology to reduce bitrate, supporting smart object recognition, motion auto-tracking, optimized image signal processing to avoid motion blur problem.
Under indoor nature light condition, and camera works under parameter settings Extra265, 1920x1080@60fps, 2048kbps, the actual measured horizontal/vertical resolution of image can reach 1100TVL, overall color reproduction is accurate, grey scale level can reach 18 grade. In our lab test, we set bitrate as VBR, camera aims at moving toy train scene, we measured the video stream’s bitrate is in the range between 1.7kbps - 67.2kbps, the average bitrate is 4.3kbps, network stream transmission is very stable, video latency is extremely low, comparing with H.265 which requires 2Mbps bitrate, the application of Extra265 has a great performance on bandwidth reduction.
Huawei M1221-F face capture box camera support 2D/3D self-adaptive noise reduction, support 120dB sensor-based wide dynamic range function. Its minimum illumination under color mode is 0.0004Lux@F1.5, under B/W mode is 0.0001Lux@F1.5, along with back-light compensation, high light compensation, AGC, auto white balance, the camera can capture high quality image whatever lighting condition is.
Completely new AI processor integrating with NPU neural network engine, with help of Huawei unique feature tracking and comparison algorithm, this face capture box camera offers exceptional functionality on face recognition. Under complex changeable scene, Huawei M1221-F face capture box camera can be applied diversity applications. It can recognize face effectively and achieve as lower as 8% repeating comparison ratio, it can definitely meet the requirement for high-density traffic places such as train station, airport, stadium, park etc.
The camera embedded 15 different kinds of smart video analytics, can support feature recognition and acquisition, even up to 10 kinds of face and human related attribute recognition, combining proximity evaluating technology, it can help you to find the right one under complex scene. For example, this camera can tell you people’s gender, age, whether wearing mask or not, people outfit. Besides, it support face recognition black list function, it can capture face then comparing it with the black list in database. It will tell you the similarity ratio and reminder you to check suspicious people if captured face match the one in black list.
In order to overcome face capturing under complex scene has high-density people and other moving objects, Huawei M1221-F face capture security camera supports T-Shot snapshot function to improve face capture efficiency, it can avoid missing target, repeat target problem. Snapshot successful rate reaches up to 96%, repeat ratio is controlled within 8%. Meanwhile, it has image filter algorithm can automatically choose the best image for comparison, can omit repeated image, on one hand to reduce the number of captured images, on the other hand to filter higher quality image helps to recognize successfully.
According to info from Huawei, the face capture security camera support intelligent analysis including fast-moving object detection, abandoned object detection, removed object detection, cross line detection, intrusion detection, area enter/exit detection, loitering detection and crowd gathering detection, image quality diagnosis, audio existence detection, abrupt audio increase or decrease detection, scene change detection, and de-focus detection. It also has dedicated algorithm including detect, track, and select dynamic face images, and generate optimal face images.
Get My Latest Posts
Subscribe to get the latest updates.
Your email address will never be shared with any 3rd parties.