OV3660 Camera Module, 3MP 60fps 69° High-Speed Sensor

SKU: FA1045-0-3
Image Sensor

OV3660 (1/5-inch CMOS)
Maximum Resolution

3MP (2048 × 1536)
Maximum Frame Rate

60fps @ VGA; 30fps @ 720p
Field of View

69° (diagonal)
Output Formats

JPEG, RAW, RGB565, YUV422/420, YCbCr422
Interface

DVP 8-bit parallel + SCCB control
Pixel Size

1.4µm × 1.4µm
Operating Voltage

Core: 1.5V; I/O: 1.8V/2.8V (module dependent)

The OV3660 Camera Module is a high-performance 3-megapixel image sensor designed for embedded vision, IoT camera projects, and microcontroller-based imaging applications. Built on OmniVision’s proven CMOS technology, this module delivers crisp image capture and smooth video performance in a compact, integration-friendly form factor.

At the heart of the module is the OV3660 sensor, a 1/5-inch CMOS image sensor capable of capturing still images at resolutions up to 2048 × 1536 pixels (3MP) and recording 720p HD video at 30fps with cropping and scaling . The sensor supports a maximum frame rate of 60fps at VGA resolution, making it suitable for applications requiring smooth motion capture . Its 69° field of view provides a balanced perspective suitable for surveillance, object recognition, and general-purpose imaging without the distortion associated with ultra-wide lenses.

The OV3660 incorporates OmniBSI backside illumination technology, which significantly improves light sensitivity compared to traditional front-side illuminated sensors. This means cleaner, brighter images in low-light conditions without requiring additional illumination hardware . The sensor also includes comprehensive image processing features built directly into the hardware: automatic exposure control, white balance, gamma correction, edge enhancement, lens shading correction, and noise cancellation — all configurable through the standard SCCB (Serial Camera Control Bus) interface .

For system integration, the OV3660 supports multiple output formats including RAW, RGB565/555/444, YUV422/420, YCbCr422, and JPEG compression . The module interfaces through a DVP (Digital Video Port) 8-bit parallel interface, making it directly compatible with ESP32, ESP32-S3, and other microcontrollers that feature native camera peripherals . JPEG compression is handled on-sensor, reducing bandwidth requirements and enabling efficient video streaming over WiFi or Ethernet.

The module is widely used in ESP32-CAM projects, AI vision development boards, and IoT surveillance systems . It is a drop-in upgrade for boards designed for the OV2640, sharing the same 24-pin DVP connector and SCCB control interface . Common development platforms include Arduino IDE (with esp32-camera library) and ESP-IDF .

Whether you are building a smart doorbell, a remote monitoring system, an AI vision prototype, or a DIY camera project, the OV3660 Camera Module offers a proven combination of resolution, low-light performance, and integration simplicity. It is suitable for individual makers, students, and businesses purchasing in bulk for product development or deployment.

Features

  • 3MP resolution (2048 × 1536) — Captures detailed still images suitable for surveillance, recognition, and documentation .

  • 60fps at VGA — Smooth video capture for motion-sensitive applications .

  • 69° field of view — Balanced perspective without wide-angle distortion, suitable for general imaging.

  • OmniBSI backside illumination — Improved low-light sensitivity without hardware modifications .

  • On-sensor JPEG compression — Reduces bandwidth and storage requirements for streaming and logging .

  • DVP 8-bit parallel interface — Direct compatibility with ESP32/ESP32-S3 camera peripherals .

  • SCCB control interface — Standard I2C-like protocol for configuring exposure, white balance, and image parameters.

  • On-chip image processing — Automatic exposure, white balance, gamma correction, and noise cancellation built in .

Technical Specifications 

# Parameter Specification
1 Image Sensor OV3660 (1/5-inch CMOS)
2 Maximum Resolution 3MP (2048 × 1536)
3 Maximum Frame Rate 60fps @ VGA; 30fps @ 720p
4 Field of View 69° (diagonal)
5 Output Formats JPEG, RAW, RGB565, YUV422/420, YCbCr422
6 Interface DVP 8-bit parallel + SCCB control
7 Pixel Size 1.4µm × 1.4µm
8 Operating Voltage Core: 1.5V; I/O: 1.8V/2.8V (module dependent)

Usage

Getting Started

  1. Connect the module — Align the 24-pin FPC connector with your development board’s camera interface. The module typically ships with an FPC cable for easy connection .

  2. Install the development environment — For ESP32/ESP32-S3, install the Arduino IDE with the esp32 board package and the esp32-camera library .

  3. Select the camera model — In the CameraWebServer example, uncomment the appropriate camera model definition. For generic boards, CAMERA_MODEL_AI_THINKER is commonly used as a starting point, though pin assignments may need adjustment .

  4. Configure resolution and format — Set pixel_format = PIXFORMAT_JPEG for streaming or PIXFORMAT_RGB565 for face detection. Choose frame_size based on your application (QVGA for high frame rate, UXGA for maximum still image resolution) .

  5. Upload and test — Compile and upload the example sketch. Open the Serial Monitor to view the IP address, then access the camera stream through a web browser on the same network .

Key Configuration Notes

  • PSRAM strongly recommended — For resolutions above VGA or high-quality JPEG, PSRAM is required to buffer frame data. The ESP32-S3 with 8MB PSRAM is an ideal companion .

  • OV3660-specific initialization — When using the OV3660, the standard initialization code may need adjustments: s->set_vflip(s, 1), s->set_brightness(s, 1), and s->set_saturation(s, -2) are commonly applied to correct orientation and color balance .

  • Pin mapping varies by board — The OV3660 uses the same 24-pin connector as the OV2640, but pin assignments differ between development boards. Always verify against your specific board’s pinout .

Typical Applications

  • IoT surveillance cameras — WiFi-enabled remote monitoring with JPEG streaming .

  • AI vision prototyping — Image capture for TensorFlow Lite, Edge Impulse, and OpenCV projects .

  • Smart doorbells and security systems — Motion detection and image capture with low-light capability.

  • Time-lapse photography — Scheduled capture for environmental monitoring or construction documentation.

  • Computer vision education — Learning image processing and embedded vision with affordable hardware.

  • OEM product development — Compact module for integrating vision into commercial devices.

Notes for Business Buyers

  • Drop-in OV2640 replacement — Same connector and interface, allowing easy upgrade of existing designs .

  • Low-light advantage — OmniBSI technology provides better native sensitivity, reducing the need for supplementary IR illumination in dim indoor environments .

  • ESP32 ecosystem compatibility — Native support through the esp32-camera library and ESP-IDF, reducing development time.

  • Suitable for bulk purchasing — Ideal for product development teams, IoT integrators, and educational institutions.

Q: What is the OV3660 Camera Module used for?

It is used for embedded imaging applications including IoT surveillance, AI vision prototyping, smart doorbells, time-lapse photography, and any project requiring image capture or video streaming with a microcontroller .

Q: Is this module compatible with the ESP32-CAM?

Yes. It is commonly used as a replacement or upgrade for the OV2640 in ESP32-CAM projects. It shares the same 24-pin DVP connector and SCCB control interface .

Q: What is the maximum resolution and frame rate?

The sensor supports up to 3MP (2048 × 1536) for still images. Video frame rates vary by resolution: 60fps at VGA, 45fps at 720p, and 30fps at 720p HD recording .

Q: What field of view does the module have?

The module provides a 69° field of view (diagonal), offering a balanced perspective suitable for most general-purpose imaging applications.

Q: Does the OV3660 support JPEG compression?

Yes. JPEG compression is handled on-sensor, which reduces the data bandwidth required for streaming and makes it ideal for WiFi-based camera projects .

Q: What interface does the module use?

It uses a DVP (Digital Video Port) 8-bit parallel interface for image data and SCCB (Serial Camera Control Bus) for configuration and control .

Q: Is PSRAM required to use this module?

PSRAM is strongly recommended for resolutions above VGA or for high-quality JPEG streaming. Without PSRAM, the frame size and quality may be limited by available internal RAM .

Q: How does the OV3660 compare to the OV2640?

The OV3660 offers higher resolution (3MP vs. 2MP), better low-light sensitivity due to OmniBSI technology, and on-sensor JPEG compression. It is a drop-in upgrade for OV2640 designs .

Q: What development environments are supported?

The module is supported by Arduino IDE (with the esp32-camera library) and ESP-IDF. It works with ESP32, ESP32-S3, and other microcontrollers featuring native DVP camera interfaces .

Q: Can I purchase in bulk for a product development project?

Yes. This camera module is suitable for both individual users and business purchasing, including product development, IoT integration, and volume production.

Related products