DAHUA NVR5232-XI/Pro 32CH 1U 2HDDs WizMind Network Video Recorder

NVR5232-XI/Pro 32CH WizMind AI NVR Specs, AI Functions & Deployment Guide NVR5232-XI/Pro 32CH WizMind AI NVR Specs, AI Functions & Deployment Guide Dahua 1U Rack-Mount 2-Bay Network Video Recorder with Xinghan Large-Scale Onboard AI, 8K HDMI Output, Full WizMind Intelligent Algorithm Support Product Overview The Dahua NVR5232-XI/Pro is a compact 1U 32-channel WizMind NVR designed for medium and large commercial surveillance scenarios, including office parks, parking lots, factory campuses, industrial workshops and residential compounds. Equipped with an industrial-grade embedded processor and independent Xinghan large-scale AI algorithm engine, this recorder supports dual intelligent computing modes: Recorder-side AI calculation and Camera-side AI analysis. It integrates face recognition, perimeter IVS, AcuPick WizSeek, SMD false alarm filtering, vehicle metadata and ANPR license plate comparison, with up to 40×1080p@30fps maximum decoding capacity and 8K ultra-high-definition HDMI local output. Core Key Points Summary: 32 IP channels input, 1U rackmount, 2 SATA HDD slots, single disk max 20TB (total 40TB storage) Max decoding: 40 × 1080p@30fps (AI disabled mode) Bandwidth split: 512/512/384 Mbps (AI off) | 200/200/200 Mbps (Xinghan AI enabled) 8K HDMI + VGA dual heterogeneous display output, independent multi-split screen Onboard AI: 8ch Face FR, 16ch Perimeter IVS, 8ch Video Metadata, 8ch AcuPick WizSeek Face database: Max 200,000 face images; ANPR supports 20,000 plate allow/block list 1. Core Hardware & Bandwidth Specifications 1.1 Basic Hardware Parameters Item Parameter Model NVR5232-XI/Pro IP Input Channels 32CH Chassis Form 1U Rack Mount HDD Bay 2 × SATA, each HDD max 20TB, total max 40TB Video Output 1 × HDMI (Max 7680×4320 8K) + 1 × VGA Max Decoding (AI Disabled) 40 × 1080p@30fps Power Supply 12 VDC, 4 A Power Consumption <15 W (without HDD) Dimension (W×D×H) 375.0 mm × 282.5 mm × 56.0 mm Operating Temperature –10 °C ~ +55 °C 1.2 Network Bandwidth Limit AI Disabled Mode: Incoming 512 Mbps / Recording 512 Mbps / Outgoing 384 Mbps Xinghan Large-Scale AI Enabled: Incoming / Recording / Outgoing all limited to 200 Mbps 1.3 Physical Port Definition Port Type Quantity & Spec Gigabit LAN 1 × RJ45 10/100/1000 Mbps USB Front USB2.0 ×1, Rear USB3.0 ×1 (backup, mouse, storage) Audio I/O RCA Audio Input ×1, RCA Audio Output ×1 Alarm I/O Alarm Input ×4, Relay Alarm Output ×2 Serial Port RS-232 ×1, RS-485 ×1 (PTZ & access control communication) 2. Complete WizMind AI Function Matrix The NVR supports two independent AI calculation modes: AI by Recorder (algorithm runs on NVR hardware, limited channel count) and AI by Camera (IPC self-computing, full 32 channels available without occupying NVR bandwidth/decoding resources). Channel capacity varies based on whether Xinghan Large-Scale AI Model is activated. 2.1 AcuPick & WizSeek Smart Target Search AcuPick (Recorder AI): 8CH (Xinghan ON) / 16CH (Xinghan OFF) AcuPick (Camera AI): All 32 channels, 36 targets captured per second WizSeek Performance: Full-channel search with AcuPick cameras; Max 8CH search for ordinary cameras; Daily target modeling up to 432,000 targets Search Supported Language: English, Traditional Chinese, Korean 2.2 Perimeter IVS Protection (Tripwire, Intrusion, Loitering) Recorder AI: 16CH (Xinghan ON) / 32CH (Xinghan OFF), 10 IVS rules per channel Camera AI: All 32 channels, 32 targets/s real-time analysis 2.3 Face Detection & Face Recognition (Access Control Integration) Face Attributes Extraction: 6 core face attributes Recorder Face Detection: 8CH (Xinghan ON) / 16CH (Xinghan OFF), 12 face images/sec per channel Hybrid FR Mode: All 32 camera-side FD + NVR unified comparison, 32 face images/s matching stream Face Database: Max 20 independent libraries, 200,000 face images total (44GB storage) 2.4 SMD Smart Motion Detection (Reduce False Alarms) Recorder SMD Plus: Disabled when Xinghan AI enabled; 32CH available with Xinghan off (filter leaves, rain, shadow interference) Camera SMD: All 32 channels, 64 targets/s analysis 2.5 Video Metadata (Human / Motor / Non-Motor Vehicle Attributes) Total 23 target attributes extracted: 9 human attributes, 9 motor vehicle attributes, 4 non-motor vehicle attributes Recorder Metadata AI: 8CH (Xinghan ON) / 16CH (Xinghan OFF) Camera Metadata AI: All 32 channels, 32 targets/s 2.6 ANPR License Plate Comparison Camera-side ANPR: Full 32 channels, 32 plate targets/s Local plate library: Store up to 20,000 plates, independent allowlist & blocklist linkage 2.7 Extra Camera-Side AI Supported (Full Channel Compatible) If matched with Dahua WizMind AI cameras, the NVR receives all intelligent data without consuming NVR computing resources: People Counting, Crowd Heat Map, Crowd Distribution, PPE Detection (Safety Helmet / Vest / Mask), Stereo Analysis, Custom Text Alarm 3. Video Decoding, Recording & Playback 3.1 Supported Input Resolution 32MP, 24MP, 16MP, 12MP, 8MP, 6MP, 5MP, 4MP, 3MP, 1080p, 720p, 960p, D1, CIF, QCIF 3.2 Decoding Capacity Comparison Resolution AI Disabled Xinghan AI Enabled 32MP@30fps 2CH 1CH 24MP@30fps 3CH 1CH 16MP@30fps 5CH 2CH 8MP@30fps 10CH 6CH 1080p@30fps 32CH 24CH 720p@30fps 40CH Max 32CH 3.2 Recording & Playback Function Record Mode: General, Motion Detection, Intelligent AI Record, Alarm Record, POS Record Max Simultaneous Playback: 16 channels synchronous playback Backup Method: USB local export / Network remote backup Playback Modes: Instant Playback, General Playback, Event Playback, Tag Playback, Smart WizSeek Playback Video Compression: Smart H.265+, H.265, Smart H.264+, H.264, MJPEG Audio Compression: G.711a, G.711u, PCM, G726 4. Alarm System & Linkage 4.1 Alarm Classification General Alarm: Motion detection, local buzzer, external alarm box, camera external I/O, network abnormal, scene change, PIR, thermal alarm Anomaly Alarm: No disk, disk failure, storage full, low quota, abnormal HDD health, network disconnection, IP/MAC conflict, fan error, video loss, tampering, camera offline, abnormal audio Intelligent AI Alarm: Face detection, face recognition, target metadata, perimeter IVS, license plate match black/white list 4.2 Alarm Linkage Actions Auto recording, panoramic snapshot, local relay output, camera external alarm trigger, audio prompt, buzzer alert, log storage, PTZ preset callout, alarm email push 5. Network, Interoperability & Access 5.1 Supported Network Protocols HTTP, HTTPS, TCP/IP, IPv4/IPv6, UDP, NTP, DHCP, DNS, SMTP, UPnP, IP Filter, FTP, DDNS, SNMP, Alarm Server, P2P, Auto Registration, ISCSI 5.2 Third-Party Camera Compatibility ONVIF, Panasonic, Sony, Axis, Arecont, Pelco, Canon, Hanwha 5.3 Remote Access Tools Mobile: DMSS App (iOS & Android), real-time preview, playback, alarm push, device management PC / Browser: Chrome, Firefox, Edge Web 5.0 interface, full configuration & footage export Interoperability: ONVIF 24.12 (Profile T/S/G), CGI, Open SDK for secondary integration 6. Certifications & Application Scenarios Certification Standard Class A CE-EMC & CE-LVD industrial certification, compliant with EN 55032, EN 62368-1 series standards, suitable for industrial factory, outdoor equipment room long-time stable operation. Typical Deployment Scenarios Commercial office buildings & business parks (face recognition access control) Indoor & outdoor parking lots (ANPR license plate recognition, vehicle metadata statistics) Manufacturing factories & workshops (PPE detection, perimeter anti-intrusion IVS) Residential compound, shopping mall, warehouse storage (AcuPick fast target retrieval, false alarm filtering via SMD)