linked to info on CSM/USGSCSM (#119)

%% Cell type:markdown id:a117baed-ab98-4499-832c-8c73a8606cc0 tags:

# Tutorial: Instantiating a CSM Camera Model from Image

%% Cell type:markdown id:3602c014-53bc-4330-a9b0-0848d4927458 tags:

Lessons learned in this tutorial:

* How to generate Image Support Data (ISD) for an image

* instantiate a CSM camera model

* Perform a simple ground to image call

Goals in this tutorial:

- [x] Generate Image Support Data (ISD) for an image

- [x] Instantiate a CSM camera model

- [x] Perform a simple ground to image call

-----

!!! info "CSM - The Community Sensor Model"

    For more info on CSM, see

    [Sensor Models](https://astrogeology.usgs.gov/docs/concepts/sensor-models/sensor-models/).

    Also see

    [Sensor Model Software: USGSCSM](https://astrogeology.usgs.gov/docs/concepts/sensor-models/sensor-model-software/#usgs-community-sensor-model-usgscsm)

    for the USGS's implementation of CSM in its software.

### 1. Prerequisites, Install Knoten

The `knoten` installation may take a little longer than usual due to the many dependencies (including ALE) involved.

%% Cell type:markdown id:ffeccab3-0d5d-4609-9c7f-871bdb69f17a tags:

```

conda install -c conda-forge knoten=0.2.1

```

%% Cell type:markdown id:faed4a43-cd06-45c7-bfa1-793978d41486 tags:

### 2. Generate an ISD from a Cube

We will use MRO data located in the `data/image_to_ground` folder containing a cube and necessary kernels for ISD (Image Support Data) generation.

*Note*: If your cube already has attached spice data, do you not have to specify kernels in the `props` param and can pass in an empty dict `{}` instead.

%% Cell type:code id:7f58cb34-d27f-456d-bfb5-f9075ca575b3 tags:

``` python

import ale

import json

import knoten

import os

# Set local data directory and paths

data_dir = '../data/image_to_ground'

cube_file = os.path.join(data_dir, 'B10_013341_1010_XN_79S172W.cub')

isd_file = os.path.join(data_dir, 'isd_file.json')

# Set local kernel paths

props = {

    'kernels': [

        os.path.join(data_dir, 'B10_013341_1010_XN_79S172W_0.bsp'),

        os.path.join(data_dir, 'B10_013341_1010_XN_79S172W_1.bsp'),

        os.path.join(data_dir, 'mro_ctx_v11.ti'),

        os.path.join(data_dir, 'mro_sc_psp_090526_090601_0_sliced_-74000.bc'),

        os.path.join(data_dir, 'mro_sc_psp_090526_090601_1_sliced_-74000.bc'),

        os.path.join(data_dir, 'mro_sclkscet_00082_65536.tsc'),

        os.path.join(data_dir, 'mro_v16.tf'),

        os.path.join(data_dir, 'naif0012.tls'),

        os.path.join(data_dir, 'pck00008.tpc')

    ]

}

# Generate the ISD string from the cube's label

isd_str = ale.loads(

    label=cube_file,

    formatter="ale",

    props=props,

    indent=2,

    verbose=False,

    only_isis_spice=False,

    only_naif_spice=False

)

# Write the ISD string to file 'isd_file.json'

with open(isd_file, "w") as file:

    file.write(isd_str)

```

%% Cell type:markdown id:4ed327aa-bffc-4316-b42f-496d9e07465e tags:

### 3. Create a Community Sensor Model

We will use Knoten's implementation of CSM as the library supports line scanner types of sensor models in the usgscsm library.

%% Cell type:code id:0c4dbf84-2986-495b-9e4a-da4c77059e7e tags:

``` python

sensor_model = knoten.csm.create_csm(isd_file, verbose=False)

```

%% Cell type:markdown id:d6973fe3-9d4a-4408-9310-50334a52ff58 tags:

### 4. Convert image coordinates into ground coordinates

%% Cell type:code id:d8f2b155-9803-4a6b-a967-bca1ef35860f tags:

``` python

# Create an image coordinate at line = 206 and sample = 206

image_coord = knoten.csmapi.ImageCoord(206, 206)

# Convert the image coordinates to ground coordinates with desired precision of 0.0

ground_coord = sensor_model.imageToGround(image_coord, 0.0)

# Output the ground coordinates

ground_coord.x, ground_coord.y, ground_coord.z

```

%% Output

    (-572485.2147483829, -79884.88742005036, -3326939.6184008163)

%% Cell type:markdown id:bf87c5a5-b26c-4168-9324-ce5b0004cc7c tags:

### 5. Convert ground coordinates into image coordinates

%% Cell type:code id:0edc0b6d-cdbe-46a8-9fdc-4ebdc4570f1a tags:

``` python

# Convert the image coordinates to ground coordinates with desired precision of 0.0

image_coord = sensor_model.groundToImage(ground_coord, 0.0)

# Output the image coordinates

image_coord.line, image_coord.samp

```

%% Output

    (205.99991086761267, 206.00000010379927)