Camera OV7670 vs. OV7620

ov7725 code programming uses
Taobao one: 7725
Information: Amoy Information Download: https://pan.baidu.com/s/1kVLrcHp
Taobao purchase

Taobao II: 7670
Profile: Http://pan.baidu.com/s/1sj8Q6Pv
Taobao purchase

OV7620 code programming uses
Taobao three: 7620
Purchase link
With no, FIFO ask customer service.
Today, there are a plethora of cameras on the market that can be applied to smart cars. In the end that is the most suitable for use in the Smart car, here Demok selected 2 typical camera OV7670 and OV7620, from its characteristics and performance angle, analyze the characteristics of the camera, for everyone to uncover the fog.
Let's go straight to the point, first of all the main differences. As we all know, OV7670 and OV7620 are all color cameras, and their photosensitive pixels are 640*480 digital cameras. They have a lot of similar parameters, but the biggest difference is that the OV7670 pixel output format is RGB565, and the OV7620 pixel output format is YUV422, this difference directly determines its position in the Smart car camera. Before we proceed with the analysis, let's start with a little bit of knowledge about the camera.

Camera trivia RGB565 and YUV422
1. What is RGB565
is known, RGB is composed of three colors of color, to display a pixel color, each pixel needs 3 byte data of R, G, B to represent, that a pair of 320* 240 of color images, you need to 320*240*3=230400b=225kb the data to store, so it appears that the space cost of storing images is enormous. In a pixel RGB, the gravity of the G component is the largest, R and B are relatively small, so people have come up with a proportional relationship with r:g:b=5:6:5 to represent a pixel, so that a pixel only needs 2 bytes to represent its color information (in this 2 byte, R accounted for 5bit,g 6bit,b of 5bit), 320*240 's color image storage space also reduced from 225KB to 150KB.
How 2.rgb565 is stored
The data for the camera is read on the rising edge of each pclk. If the camera default output format is RGB565, then the camera after power up, each trigger 2 pclk, read 2 bytes, is a pixel. In this pixel, the RGB distribution is shown in the following figure: In the first byte, y[7..3] is r[4..0],y[2..0] g[5..3]; in the second byte (Second byte), y[7..5] is g[2..0],y[ 4..0] is b[4..0]
2. What is Yuv4:2:2
People's eyes are more sensitive to low-frequency signals than to high-frequency signals, and in fact, the human eye is more sensitive to visual changes than to color changes. Therefore, the RGB three-color signal is changed to YUV to indicate, where Y is grayscale, UV is chromatic aberration. In the same way that a color image is represented, the YUV444 is a lossless storage method, but requires 3 bytes and has a large storage space overhead. Because the Y component is more important than the UV component, people use YUV422 to represent it. So that the image is compressed a lot, one byte can represent its color information.

Return to the contrast of OV7670 and OV7620.
Just from childhood knowledge, understand that OV7670 is RGB output, and OV7620 is YUV output, although theoretically understood, but we may not know the actual corresponding camera is how to output.
For OV7670, because it has only a set of parallel data ports y[7:0], all pixel data is out of this data port, so each time the data port is read, one byte (byte) can be read. The following figure shows the format of the K-pixel (2K bytes) output.
For OV7620, it has 2 sets of parallel data ports y[7..0] and uv[7..0], where for the data port y[7..0], the output is the grayscale value y, for uv[7..0] output chroma signal UV. The following figure shows the format of the K-pixel (k-byte) output.
The above 2 diagram has been very clear explanation of the output format of the differences, and now the pros and cons of the summary.
For the Smart car race, the track is a white black line, so we only care about the grayscale value of the image, not need their color value. Thus, for OV7670, it has only a set of data ports, the default output RGB color image, not for us, and for OV7620, it has two sets of data ports, a set of data port is the gray value of y, we do not need to pipe the output of the UV pin signal, only the Y port data, will be able to perfectly reflect the information of the track. For the all-white track background, the collected data is 255, for the black track, collected back data is 0, so that can be very good differences between the opening and the background.
This can be summed up, OV7620 more suitable for Smart car camera.

The difference between a camera without a FIFO and a FIFO camera:

The camera with FIFO is more than 3M cache, the data collected can be temporarily present in this cache, and the image data in the cache can be read, so the MCU speed requirement is reduced when the image data is collected by single chip microcomputer. (Routines store image data in RGB565 mode)
CMOS camera compared to other sensors, it is difficult, involving a lot of things, please before purchasing and use, do more evaluation and research, and then decide
OV7670 FIFO camera, the image sensor and image buffer device al422b organic combination, solve the low-end single-chip computer image acquisition speed bottleneck problem. Basic principles of use as shown:
Users only need to read the data in the FIFO sequence, control the relevant pin, can be directly from the MCU IO Port reading data, the data into memory or into the screen memory display, can be controlled by low-speed MCU operation, and can be simple processing of data, such as black and white recognition.
FIFO Camera factory detection, the purpose is to ensure quality, to prevent problems to the two sides to bring a variety of trouble.
Introduction of OV7670 with FIFO module
1. Introduction:
The OV7670 with FIFO module is a module with buffer storage space for slow MCU to realize image acquisition control. This module adds a FIFO (first-in, in-out) memory chip, also contains 30w pixels CMOS image sensor, 3.6mm focal length lens and lens holder, on-board CMOS chip needs a variety of different power supply (power requirements are described in the chip data file), the Board also leads to control pins and data pins, Easy to operate and use.

As shown in the figure, the pins required to control the sensor are defined as follows:

3v3-– Input supply voltage (recommended to use 3.3,5V, but not recommended)

gdn-– Pick-up point

Control clock of the SIO_C-SCCB interface (note: Some low microcontroller requires pull-up control, similar to the I²c interface)

Serial data input (out) end of the SIO_D-SCCB interface (note: Some low microcontroller requires pull-up control, similar to the I²c interface)

vsync-frame sync signal (output signal)

href--line synchronous signal (output signal, generally can not be applied, special case use)

d0-d7-data port (output signal)

reste-Reset port (normal use pull high)

pwdn--Power selection mode (normal use pull low)

strobe-Camera Flash Control port (normal use can not be required)

Fifo_rck-fifo memory Read Clock control end

Fifo_wr_ctr--fifo Write control end (1 for Allow CMOS write to fifo,0 to prohibit)

Fifo_oe--fifo shut-off control

FIFO_WRST-FIFO Write pointer reset end

FIFO_RRST-FIFO read pointer reset end

3. Control mode description

As the use of FIFO as a data buffer, a simple data collection, users only need to care about how to read, do not need to care about how the specific data collected, so as to reduce or even do not care about the CMOS control and timing relationship, you can achieve the image acquisition.