Data Realization: When Pix4d makes data, there is an option of "merging data" by default, so the image terrain made by pix4D includes at least two copies, and the default is three copies (transparent and opaque). Transparency means that places where there is no data on the edge will be displayed in a transparent way, and the amount of data will become larger because there is one more transparent channel. Blocking refers to whether the final data of production is a single tif or multiple TIFs. By default, both exist.
There are fewer Smart3D options, and only one kind of data is the result of blocking.
The result form of data is mostly tif, and the image and terrain suffix are in tif format, but the data storage is different, one is 24-bit true color and the other is 8-bit gray scale.
Data magnitude: Generally, the image data with a resolution of 2 cm and an area of 1 km2 can reach more than 3Gb (most of which are compressed by tif itself), and the terrain data will be slightly smaller.
I tested it once before. It takes more than 5 minutes to open a single tif of 10GB. If the machine configuration is not particularly good, it often cannot be opened.
The method mentions high efficiency: traditional GIS giants such as ArcGis and SuperMap also deal with big data by creating tif pyramids. This process is very slow, and the operation is complicated, the failure rate is still relatively high, and the requirements for computer configuration are also high. General GIS white is more difficult to get (many people who know about UAV data production don't even know the existence of this method). Incidentally, GlobalMap is much more efficient than the above software in opening tif big data. Conventionally, through this method, the data opening efficiency can be improved by several times to dozens of times.
The second method to improve efficiency: data preprocessing, which processes tif into tile format that can be loaded quickly, so that browsing data is like browsing online maps, and the data is loaded dynamically. Many offline navigation packages are based on this principle. In this way, the data browsing speed is basically seconds (milliseconds).
LSV supports both methods. For the first method, only data browsing after pyramid creation is supported, and the process of pyramid creation is not supported. The second method, lsv not only provides data processing methods, but also supports data merging and data batch processing.
Please refer to its official help document: 10. 1 image processing (offline data production) for the specific operation process.
The first stage of data appeal, most of them are watching. If you can see it, 80% of the problems will be solved. What do you think? How do you look better?
The basic situation of data production is:
1. The aerial photography of an area is carried out in a unified form, and the data resolution is consistent.
2 The overall situation is unified aerial photography, and aerial photography with higher accuracy is carried out in some areas (for example, the overall accuracy is 10cm, and the core area is 2cm).
3. Data in multiple areas need to be quickly switched and viewed centrally.
The data of the same area are regularly produced by aerial photography, with historical data.
5 The data is opaque when it is made, and there are white edges or black edges after loading.
The data is directly provided to Party A. Party A doesn't care about the data format or the data form, but only wants to open the software to read the data when it wants to.
Solution:
1. Layer management using lsv. For the processed tif data (the processing result is lrp), it is directly added to the layer. Every time the software opens the data, it is there (as long as the metadata has not been deleted or the data location has not been modified).
2. Layer order: For high-precision data mixing, high-precision data can be on top and low-precision data can be on the bottom. You don't need to pay attention when browsing, just zoom in and the software will automatically display the clearest data.
3. For white edges or black edges, LSV can process them when stitching images.
4. Multi-regional data can be merged into one data, with splicing function for overall control and timely disconnection.
5. For historical data, you can quickly check the data of different periods by checking and unchecking layers (standardization of data names is recommended).
6. For Party A who is technically a little white, you can directly configure it for him, and he only needs to start the inspection.
7. If you want to be cool, the image and terrain are superimposed and displayed, which is similar to the tilt effect, but the texture of the facade effect is worse than that of the tilt model. However, the immersion effect of LSV is very strong in areas with large topographic relief.