Red square on black screen - raspberrypi_video

[Henrikcox]

I've been trying to set up a Lepton Thermal Camera (breakout v1.4) with both a raspberry pi 3 and a zero W. I started with this wiring and setup guide and have been troubleshooting with all the suggestions mentioned in this thread with no luck.

Running sudo ./raspberrypi_video

• Black pop up window with red square in the upper left corner. 'Perform FCC' button works and causes the Lepton camera's shutter to close and open.

Running sudo ./raspberrypi_capture [with output file name]

• Message 'spi mode: 0 bits per word: 8 max speed: 16000000 Hz (16000KHz)', terminal hangs on a blank command line. Ctrl-c to quit

Running sudo ./raspberrypi_overlay [with options such as '-a 155']

• terminal hangs on a blank command line (pi camera IS connected for this one)

Running sudo ./raspberrypi_qt

• All-red window pops up, 'snapshot' creates and rgb#.jpeg file, error message in terminal 'Error transferring SPI packet'

To troubleshoot I have:

• Carefully removed and put back the Lepton module from its breakout board.
• Placed the CS pin in #24 (CE0) and #26 (CE1) and matched it correspondingly for spidev in the LeptonThread.cpp script (0.0 for CE0, 0.1 for CE1)
• From an answer in this thread, the wiring guide I've followed is allegedly false. So I experimented my full pin layout between these two patterns:
  • [SPI0] MOSI: #19, MISO: #21, SCLK: #23, CE0: #24, CE1: #26
  • [SPI1] MOSI: #38, MISO: #35, SCLK: #40, CE0: #12, CE1: #11
• Set DEBUG_LEPTON to true in the Lepton_thread.h script, then back to false but commented out the debug defines surrounding line 48, and changed 'sequence' to 'sequencex'
• Changed the SPI speed to 18000000 from 10000000
• Checked the voltage through the Lepton breakout board while plugged in to make sure it isn't fried (its shutter also goes off every few mins).

In addition to LeptonModule (C/C++) I have also tried the same with pylepton (Python) with identical results.

Any insights and advice are appreciated in advance

[hanifizzudinrahman]

thx for the solution after this, i will try it...

[hanifizzudinrahman]

btw, i want to ask i have done with

• Carefully removed and put back the Lepton module from its breakout board. (sometimes it doesn't run and sometimes it runs)
• Placed the CS pin in #24 (CE0)
• [SPI0] MOSI: #19, MISO: #21, SCLK: #23, CE0: #24,
• Changed the SPI speed to 18000000 from 10000000 (in SPI.cpp) unsigned int spi_speed = 18000000; and spi_speed = 18000000;

in yours, we must Set DEBUG_LEPTON to true in the Lepton_thread.h script, then back to false but commented out the debug defines surrounding line 48, and changed 'sequence' to 'sequencex'

but, in mine there is no Lepton_thread.h but LeptonThread.h and the code is

#include "LeptonThread.h"

#include "Palettes.h" 
#include "SPI.h"
#include "Lepton_I2C.h"

#define PACKET_SIZE 164
#define PACKET_SIZE_UINT16 (PACKET_SIZE/2)
#define PACKETS_PER_FRAME 60
#define FRAME_SIZE_UINT16 (PACKET_SIZE_UINT16*PACKETS_PER_FRAME)
#define FPS 27;

LeptonThread::LeptonThread() : QThread()
{
}

LeptonThread::~LeptonThread() {
}

void LeptonThread::run()
{
    //create the initial image
    myImage = QImage(80, 60, QImage::Format_RGB888);

    //open spi port
    SpiOpenPort(0);

    while(true) {

        //read data packets from lepton over SPI
        int resets = 0;
        for(int j=0;j<PACKETS_PER_FRAME;j++) {
            //if it's a drop packet, reset j to 0, set to -1 so he'll be at 0 again loop
            read(spi_cs0_fd, result+sizeof(uint8_t)*PACKET_SIZE*j, sizeof(uint8_t)*PACKET_SIZE);
            int packetNumber = result[j*PACKET_SIZE+1];
            if(packetNumber != j) {
                j = -1;
                resets += 1;
                usleep(1000);
                //Note: we've selected 750 resets as an arbitrary limit, since there should never be 750 "null" packets between two valid transmissions at the current poll rate
                //By polling faster, developers may easily exceed this count, and the down period between frames may then be flagged as a loss of sync
                if(resets == 750) {
                    SpiClosePort(0);
                    usleep(750000);
                    SpiOpenPort(0);
                }
            }
        }
        if(resets >= 30) {
            qDebug() << "done reading, resets: " << resets;
        }

        frameBuffer = (uint16_t *)result;
        int row, column;
        uint16_t value;
        uint16_t minValue = 65535;
        uint16_t maxValue = 0;

        for(int i=0;i<FRAME_SIZE_UINT16;i++) {
            //skip the first 2 uint16_t's of every packet, they're 4 header bytes
            if(i % PACKET_SIZE_UINT16 < 2) {
                continue;
            }

            //flip the MSB and LSB at the last second
            int temp = result[i*2];
            result[i*2] = result[i*2+1];
            result[i*2+1] = temp;

            value = frameBuffer[i];
            if(value > maxValue) {
                maxValue = value;
            }
            if(value < minValue) {
                minValue = value;
            }
            column = i % PACKET_SIZE_UINT16 - 2;
            row = i / PACKET_SIZE_UINT16 ;
        }

        float diff = maxValue - minValue;
        float scale = 255/diff;
        QRgb color;
        for(int i=0;i<FRAME_SIZE_UINT16;i++) {
            if(i % PACKET_SIZE_UINT16 < 2) {
                continue;
            }
            value = (frameBuffer[i] - minValue) * scale;
            const int *colormap = colormap_ironblack;
            color = qRgb(colormap[3*value], colormap[3*value+1], colormap[3*value+2]);
            column = (i % PACKET_SIZE_UINT16 ) - 2;
            row = i / PACKET_SIZE_UINT16;
            myImage.setPixel(column, row, color);
        }

        //lets emit the signal for update
        emit updateImage(myImage);

    }

    //finally, close SPI port just bcuz
    SpiClosePort(0);
}

void LeptonThread::performFFC() {
    //perform FFC
    lepton_perform_ffc();
}

and also, there are no **DEBUG_LEPTON**  and 
**'sequence'**
 or 
**'sequencex'**

[Henrikcox]

The LeptonThread.h file I'm referring to is in the raspberrypi_qt folder (under software). You have the wrong file open.

Follow the third comment's example in this thread. Hopefully that works for you; I still have not been able to get mine working.

[hanifizzudinrahman]

i have try it but still wrong when try raspberrypi_qt the message error is "Error transfering SPI Packet" i have modified the code in LeptonThread.cpp

• const char *LeptonThread::device = "/dev/spidev0.0"; so, im using [SPI0] MOSI: #19, MISO: #21, SCLK: #23, CE0: #24,
• unsigned int LeptonThread::speed = 18000000; before the speed is 16000000 in LeptonThread.h
• std::list< std::pair<int, int> > sequencex; before is sequence

[Henrikcox]

Yes that is one of the errors I'm getting right now too. Still have not found a solution to it I'm afraid.

[hanifizzudinrahman]

i use a new code from Frank

LEPTON 2.zip

I will say full thank you to you because, with this code, is absolutely make me happiest person in the world, hehehe...

But, there is little problem in my project (Lepton 2.5 with board V1.4 + Raspberry Pi 3 B) my schematic is

• CS = SPI0 CS0 (Pin 24) # Use spidev0.0
• MOSI = SPI0 MOSI (Pin 19)
• MISO = SPI0 MISO (Pin 21)
• CLK = SPI0 SCLK (Pin 23)
• GND = GND (Pin 14)
• VIN = 3.3V PWR (Pin 17)
• SDA = I2C1 SDA (Pin 3)
• SCL = I2C1 SCL (Pin 5)

According with the suggestion above, i make some change in ....

• LeptonThread.cpp
• const char *LeptonThread::device = "/dev/spidev0.0"; # (can change in spidev0.1)
• unsigned int LeptonThread::speed = 18000000; # (from 16000000)
• qDebug() << resets << "reset," << errors << "errors"; # (before this line is commented)
• LeptonThread.h

• define DEBUG_LEPTON true # (before this line is false)

when i run the Lepton-CaptureRaw, the result is good like this, but...

but..., sometimes when i try 10 times run this i can video it the thermal video and when i press the button "Snapsot" it will give me the .jpg picture (1-rgb.jpg, 2-rgb.jpg) and .bin format (1-raw.bin, 2-raw.bin) that this raw format will show a temperature value (i'm not yet with this) but 2 times will fail. The error is "Error transfering SPI packet" and the layer full of red. Can you help me with this? why this happen?

[Henrikcox]

I'm glad you're making progress with it. Is the LEPTON 2.zip file the same as the our_video.zip file from the top of that thread (https://groups.google.com/forum/#!topic/flir-lepton/GWHD1KMVYaE)?

I was following the same thing as you did but still get the red screen and either "Error transferring SPI packet" or "Packet reset counter hit 750." But I am running raspberrypi_video with sudo ./raspberrypi_video

Are you running a different file, how are you running it? I am not seeing where Lepton-CaptureRaw that you're running is.

[gholibeigian]

Hi, It seems very crazy, but when I changed the jumper wires with the short ones(shortest), I did not get this error anymore!